Canadian Journal of Communication Vol 39 (2014) 539–556
©2014 Canadian Journal of Communication Corporation

Innovation and Compliance in Making and Perceiving the Scientific Visualization of Viruses

Roberta Buiani
The Fields Institute for Research in Mathematical Sciences, University of Toronto

Roberta Buiani, PhD, is co-founder of the ArtSci Salon at the Fields Institute for Research in Mathematical Sciences, University of Toronto, and program advisor for the Subtle Technologies Festival, Toronto. Email: .

ABSTRACT This article investigates the variety that typifies visualizations of the H5N1 virus. It argues for virus visualizations to be examined as products of processes, instruments, and socio-cultural assumptions, rather than as mere illustrations. Visualization unfolds through a series of tensions between two contending forces. The first implements normative rules and cultural narratives. The second seeks to escape them. This double tendency recurs at every stage of the visualization process. It is only by considering all stages of visualization that one can understand both its complex socio-technical articulation and the concerted role played by laboratories, graphic studios, and the general public in modulating and co-producing these tensions.

KEYWORDS  Viruses; Electron microscopy; Scientific visualization; Biopolitics; Biopower

RÉSUMÉ  Dans cet article j’interroge la diversité qui caractérise les pratiques de visualisation en utilisant à titre d’exemple le virus H5N1. J’invite à examiner ces images comme étant les produits des processus, des instruments, et des postulats socio-culturels, plutôt que comme de simples illustrations. La visualisation se déroule à travers des tensions entre deux forces qui s’exercent dans des directions différentes: la première met en œuvre des règles et des narrations ancrées dans la culture; la seconde cherche à y échapper. Cette double tension se reproduit à chaque étape de la visualisation. C’est seulement en considérant toutes les étapes de ce processus de représentation que l’on peut saisir la complexité de son articulation socio-technique de même que le rôle joué par le laboratoire, le studio graphique, et le public dans la modulation et la co-production de ces tensions.

MOTS CLÉS  Virus; Microscopie électronique; Visualisation scientifique; Biopolitique; Biopouvoir


Thanks to a variety of scientific processes and visualization instruments (e.g., electron microscopy, digitization, visualization software), viruses can now be turned into visual artefacts for the enjoyment of a general and professional public. Molecular models, microscopy scans, and other colourful images claiming to visually reproduce particular categories of viruses grace the pages of popular science magazines and peer-reviewed journals. The general public largely accepts these images as products derived from direct scrutiny of viruses by means of high-resolution devices (Pauwels, 2006, p. viii). Unaware of the laborious operations and processes required to make these images possible, the typical assumption is that what appears in front of one’s eyes is simply a magnified reproduction of a particular virus.

The illustration of the H5N1 virus that appeared a few years ago on a special issue of the New Scientist dedicated to the bird flu pandemic (Figure 1) is a case in point. Occasional readers may not associate such refined and detailed images with repertoire images and electron-microscopic scans (Figure 2), or with the myriad artistic renditions and illustrations of the virus currently circulating across the media spectrum. This is equally so for images of other viruses (e.g., Ebola, Hepatitis, HIV, SARS). The heterogeneity characterizing the visualizations used to illustrate each virus complicates its identification for a general audience. This raises questions about the extent to which arbitrary aesthetics, scientific or technical limitations, and/or deceitful instrumentalization are involved in the realization of these images. It also fosters the impression that no definite answers or clear explanations exist to help redress the halo of mystery and the fear viruses generate. This, in turn, serves to confirm and even exacerbate anxiety about a substance whose visual rendition encapsulates all the risks associated with infectious diseases.

Figure 1


Figure 2


Using reproductions of the H5N1 virus as examples, this article considers the products of visualization displayed in science periodicals as the culmination of a series of processes spanning from the ability (or obligation) to isolate a viral molecule, to the processing of biological or digital data, to the operations needed (e.g., freezing, coating, digitizing, data mining, processing) to achieve a satisfactory visual product for public and scientific consumption (Bozzola & Russell, 1999). It treats each stage of visualization as the locus of biopolitical tensions that simultaneously implement and transcend default scientific and cultural regimes of knowledge associated with the management and dissemination of virus-related information. These tensions solicit innovation while binding visualizations to scientific mores, equally obstructing and propelling the unbridled and somehow un-ordered development of the practice of visualization. In so doing, they speak to the relations between laboratory science, graphic design, and the general public.

In explicating biopolitical tensions, the discussion below focuses on practices of visualization occurring in laboratories and beyond, performed by experts in the field for a non-expert public. Particular attention is given to the significance and co-responsibility of material—both natural and instrumental—and to politico-cultural variables influencing the development of these activities. The first part of the discussion examines the overlapping technical and scientific processes involved in capturing and preparing viruses for the production of visualizations. This is followed by an analysis of the ambiguous distinction between the technical elements needed to produce legible images, and the stereotypical perceptions they help to disseminate. The article concludes with an examination of how the increasing specialization and fragmentation of the industry engendered by technologies of visualization perpetuates tensions between innovation and tradition and blurs the boundaries between the professional and the public.

Navigating visualization

The practice of visualization is not comprised of merely aesthetic choices. Human elements and instrumental conditions contribute equally to this activity, with scientists’ decisions, public perceptions, and the techno-aesthetic interpretations of graphic designers intersecting with the quality and state of the microscopes used, the software employed, and the preparation techniques applied. The particular molecular nature and submicroscopic size of viruses further hinders the ability to capture and classify these substances according to definite taxonomic categories. Likewise, cultural and popular assumptions defining the perniciousness of viruses tend to affect the ways in which they are visually reproduced.

From the preliminary data collection and imaging processes conducted in laboratories, to the graphic re-elaboration for public and academic display, visualization is the result of a dynamic intersection of discursive practice and the material world (Barad, 2007). As Myers (2014) suggests, the act of visualizing is an act of rendering. It is not just a matter of reproducing an “object that can stand in for something else” (p. 154), but it is a combination of gestures, performative by nature, referring not just to “the object that is rendered, but also to the subject, the one who renders, and the activity of rendering” in order to “pass an object or communication from one person to another” (p. 154).

All intersecting technical and scientific aspects of visualization, including expectations and tacit rules, shape and are shaped by the visual products emerging at the end of the process (Ruivenkamp & Rip, 2014). “The laboratory,” as Coopmans (2014a) explains, “extends to other spaces and places via collaborative ventures, shared data centres, and information and communication technologies,” challenging “the very distinction between laboratory and field” (p. 3). This acknowledgement characterizes scholarship in science and technology studies (STS) devoted to expanding research beyond the enclosed space of the laboratory environment. Inspired by Latour’s (1988; 1996) actor network theory and Latour and Woolgar’s (1979) analysis of the rich activity and relationships interwoven and performed in the laboratory, STS researchers have increasingly understood this space as composite and multi-layered, containing a range of instruments, protocols, processes, and power relations (Latour & Woolgar, 1979; Latour, 1988; 1996).

Feminist science studies contributed to refining this model, demonstrating that science and its mechanisms of representation are operations best understood as material-semiotic enactments (Haraway, 1997; Stengers, 2010) or entanglements of matter and meaning (Barad, 2007). A recent wave of studies focusing on representation in scientific practice—also known as RiSP, and exemplified by two seminal volumes published in 1990 and 2014 (see Coopmans, 2014a, 2014b)—has shifted attention to ontological and semiotic continuities existing between diverse scientific practices. Those working in this domain deem such an approach necessary for examining scientific practices that increasingly rely on digital technologies and image processing—including practices that manipulate both wet and digitized material—because they expand the realm of activity beyond the laboratory.

Some important questions in this regard include: What happens when visualizations reach the public? How does visualization reflect cultural fears and anxiety toward infectious diseases? How are these reactions influencing the work of scientists and graphic designers? While such questions are rarely interrogated directly by STS researchers, they are of particular interest to those working in the fields of media and communication studies. The central premise underlying my argument is that special consideration needs to be given to the role of the public both in perceiving and understanding and in disseminating and suggesting the forms, styles, and aesthetics of virus visualizations.

As Ruivenkamp and Rip (2014) suggest, images of substances like viruses that are not visible to the human eye, and which are “assumed to be ‘down there’ based on data provided by instruments” (p. 177), are “guided by expectations about what can be ‘seen’ [at this scale] and rules [about] how [this scale] should be visualized” (p. 179). Scientists are aware of the potential effects that visualization can have on publics who consume these images, in part because they are themselves exposed to the same popular cultural narratives and anxieties surrounding viruses. Decisions about the appearance to be given to the molecular or superficial composition of these substances are based on parameters established by teams of experts who draw from conservative visual conventions to mark a continuation from, and a belonging to, a certain scientific tradition (Rasmussen, 1999). However, these experts also draw on popular iconography that partially satisfies the expectations of the public regarding particular substances, facilitating their identification and prolonging their popular reputation (Ruivenkamp & Rip, 2014). It is possible to evince the unfolding of this tension by performing a semiotic analysis of H5N1 illustrations disseminated by media outlets. In this case, new software combined with advanced mapping and visual techniques manifest colours and composition styles associated with well-established visual and technical conventions for “consensual ‘seeing’ and ‘knowing’” (Lynch, 1988) that obey the “disciplinary and institutional frameworks” (Mody, 2014) of scientific practice while sustaining inherent cultural assumptions about the virus.

The (bio)politics of viral representation

Despite the sophisticated instruments available today, the size and the nature of viruses (i.e., their material aspects) escape the formulation of standardized visualization models and methods of analysis. As a result, the visual features of scientific visualization have to be constantly adjusted to accommodate newer information about viruses and the diseases they sometimes cause; different circumstances and research questions require different types of visualizations (Lynch, 2006a). A variety of instruments and protocols strive to obtain a clear definition of viruses as distinct and self-contained substances (Lynch, 2006b). Typically, these practices domesticate viruses by turning them into definite objects (van Loon, 2002a; 2002b) and by ordering them according to variety, genus, and species, thereby associating them to given categories with the help of established and readily available research practices. As van Regenmortel and Mahy (2004) explain:

When a virus undergoes its so-called life cycle, it takes on various forms and manifestations, for instance, as a replicating nucleic acid in the host cell or vector. One stage in this cycle is the virus particle or virion, which is characterized by intrinsic properties such as size, mass, chemical composition, nucleotide sequence of the genome, and amino acid sequence of protein subunits, among others. (p. 8)

Viruses can be mainly studied and classified in their virion state—that is, when they reach a particular phase of their development. As their existence can be verified and studied mostly in relation to their host, and not as separate entities, viruses are classified according to rules that rank them in a specific moment of their development.

Despite these rules, the juxtaposition of different technologies and techniques needed to extract information from viruses produces diversified visual results that may partially invalidate established guidelines (Steinman & Steinman, 2011). The difficulty in deciphering viruses in their material configuration, combined with the ways in which scientists and publics imagine them, fosters great diversification in visualization, as different attempts to isolate and visually present these submicroscopic substances prioritize particular views or aspects over others.

The notions of biopower and biopolitics can be used to explicate how these apparently contradictory aspects may coexist in this practice. Roberto Esposito (2008) summarizes the two concepts in the following questions: “In what sense does life govern politics, or in what sense does politics govern life? Does it concern the governing of or over life?” (p. 15) Although often used interchangeably to describe the same phenomenon, the difference between biopower and biopolitics lies in whether one speaks of governance as the subjection of life to politics (e.g., a substance, the virus, acquiring life properties when in collision with its host), or whether we construct politics in the name of life. The visualization of viruses exemplifies and intersects with both instances. At one level, scientific visualization attempts to govern viruses by containing them within given categories and standardized visual forms. At another level, the process of visualisation embodies the complications arising from governing viruses, insofar as technical and discursive obligations overlap with different approaches to, and perceptions of, these submicroscopic substances. These tensions reflect the unfolding of a never-ending conflict between the “life” of the submicroscopic substance and the “politics” of production and representation (Esposito, 2010, p. 28).

Understood as the locus of convergence of many processes, scientific visualization manifests itself as the product of biopower and biopolitics, raising issues at material and cultural levels. This includes, on the one hand, the management of viruses within a disciplinary realm and their visual containment as well-identifiable and recognizable substances. On the other hand, it emphasizes the conflicts arising from attempts to analyze a volatile substance and to master a set of instruments that may produce diverse outcomes (Esposito, 2008). Biopower and biopolitics are fused when popular images of viruses equally encourage the viewer to take part in biosocial reactions and to share pre-constituted ideas regarding infectious diseases (Rabinow, 1992), or when they make one question the messages and the accuracy of these same images.

Foucault (2003) defines biopower as the most recent power configuration in a three-stage system: first, a sovereign and juridical power; second, a power based on “disciplinary mechanisms”; and third, a system whose “purpose is not to modify any given phenomenon as such, or to modify a given individual insofar as he is an individual, but, essentially, to intervene at the level at which these general phenomena are determined, to intervene at the level of their generality” (p. 247). This regime of power makes sure that this system maintains a balance by anticipating any surprise. Unlike sovereignty, and unlike the power to “take life and let live,” biopower is the “power to intervene to make live” (p. 248), that is, to regulate the lives of those who are alive. Biopower no longer deals with the legal subject over whom a sovereign holds the power of life and death, but seeks to achieve a mastery at the level of life itself (Macey, 2009), “taking control of life and the biological processes of man-as-species and … ensuring that they are not disciplined, but regularized” (Foucault, 2003, p. 246).

The mechanisms introduced by biopower are much more subtle than those of previous regimes of power. They are not directly coercive, operating instead by measuring general trends, mapping tendencies, extrapolating phenomena and, ultimately, releasing recommendations about the way populations ought to behave in order to avoid unexpected disruptions to the system. Given its mandate to create a system of control and to provide data and visual products that are regulated, predictable and “typical,” visualization becomes a mechanism of governance of the virus based on biopower. By capturing and measuring the molecular structure and monitoring the behaviour of viruses through statistical estimates, forecasts, and approximations, visualization makes decisions based on average characteristics, rather than on single behaviours and/or unique structures (Flegler, Jr., Heckman, & Klomparens, 1997). Templates for illustrations are, in fact, based on molecular data and on models that can be downloaded from the Protein Databank (see Protein Data Base, n.d.), which provides numerous examples to draw from, to match, and to compare with experimental data. Thus, the viruses we “see” in popular science illustrations often represent a substance in its typical, rather than its unique, state (Chatterjee, Roy, Laskar, & Swarnakar, 2013).

The tendency to produce and disseminate images that are interpreted as reliable reproductions of viruses speaks to the often unquestioned power and authority of science and technology. This power extends over the biological specimen that visualization claims to reproduce despite the technological layers juxtaposed in turning the data retrieved from the microscope into a visible object (van Loon, 2002a). Questions arise, therefore, about the goals and motivations of the product, as well as its veracity. Yet a general confidence in technologically enhanced objects seemingly dissuades many from interrogating the manipulative and instrumentalizing potentials of the processed image (van Dijck, 2005). The presence in these images, which satisfy basic cultural narratives about viruses, contributes to their widespread uncritical acceptance. The scientific desire to isolate, classify, and control the visual appearance of viruses reflects the principles of a forecast-and-control society driven by the necessity to deal with standardized, easily replicable, and predictable outcomes (van Loon, 2002b). In this instance, biopower manifests as an imperative to govern a specific disciplinary area that relies on fast evolving digital imaging and laboratory hardware (Coopmans, 2014b), by homogenizing and regulating its processes and imagery. This, however, clashes with an opposing need to keep up with new findings and methods for analyzing and displaying viruses in increasingly accurate details.

Parallels can be drawn between the way in which Foucault (2007) describes the relation between security and population and the relation that exists between visualization and viruses. Like security, whose aim is to study a phenomenon in order to anticipate its behaviour and minimize any adverse surprise it may cause (Foucault, 2007; Thacker, 2009; van Loon, 2002b), visualization monitors and studies the structure of viruses in order to understand—and thus predict—their chemical configuration and behavioural patterns. Moreover, visualization does not originate from a specific set of policies or from a single institution imposing a fixed agenda on how visual products should look or what they should communicate. It is a product of the collaborative work of scientific study groups, laboratory teams, graphic designers, chief scientists, graduate students, and other stakeholders (Chandler & Roberson, 2009).

Visualization is also tied to aesthetic and cultural norms that cannot be ignored when trying to reproduce the appearance of viruses. Two complementary questions arising in this context are: how can scientific information about viruses be interpreted and/or discerned uniformly? How can predictions be made when the object of research is not a fixed one? With mutating objectives tied to different contingencies, and with arbitrary forms characterizing visualization, inconsistencies are unavoidable. These factors contribute to the diversification of virus visualization. In addition to the obligation to regulate the form and aesthetics of virus illustrations and the desire to innovate, transform, and modify the way in which they appear visually, scientific visualization expresses difficulties pertaining to the governing of the life of viruses, and therefore to the biopolitics of visualization. Put simply, mechanistic and conventional methods of display and aesthetics fail to grasp the significance and the complexities of submicroscopic entities like viruses.

Following Esposito’s (2008) analysis of biopolitics, “the terms from which biopolitics is formed (life and politics) cannot be articulated except through a modality that simultaneously juxtaposes them” (p. 32). The juxtaposition of bíos (i.e., life, conceived here in general terms) and politics can be found in the seemingly differing goals and scientific questions that visualization is simultaneously supporting. It emerges from the incongruity between seeking to tame virus visualization, on the one hand, and freely representing viruses, on the other hand. In the words of Esposito, “either life holds politics back, pinning it to its impassable natural limit, or, on the contrary, it is life that is captured and prey to a politics that strains to imprison its innovative potential” (p. 33). Biopolitics, Thacker (2009) adds, “becomes the governance of vital forces” (p. 137). It is an attempt to understand and pin down something that cannot be fully assimilated because of its own shifting nature and the many forces and pressures shaping, modifying, and converging into this object and, ultimately, its visual display.

According to this interpretation, the rapid proliferation and variety of scientific visualization is but one material manifestation, rather than the definite product of struggles characterizing tensions between bíos and politics. Scientific visualization engages with incessant innovation, appearing unable to reach agreement or fixed regulations about how viruses should be represented. Looking at the processes comprising visualization and its quest to produce the images found in science periodicals offers clues about its management over and of viruses. It also reveals how observers are encouraged to accept typical figurations of viruses that, in turn, shape, by way of cultural tropes and collective narratives, how they are displayed.

The nature of the submicroscopic: A multi-part process

Visualization facilitates comprehending the immaterial and the invisible. However, isolating and studying viruses is a challenge that manifests itself at practical and perceived levels. The heterogeneity produced by the technology required to illustrate viruses has to be negotiated with expected visual patterns. As Stengers (2010) points out, “constraints, requirements and obligation operate in a way that is detached from the specific case of experimental invention” (p. 52). The varied results obtained from the subjection of viruses to the electron microscope are examined, compared to similar experiments, and selected according to specific objectives (Chandler & Roberson, 2009). The processes leading to visualization isolate, objectify, and idealize viruses in order to fit specific goals satisfying the constraints dictated by scientific discourse. In other words, preparatory procedures are practised with the intention of turning the substance of investigation into a “docile object … fit to be studied according to the established methods and mores of science, the instrumentations and the laboratory set-up” (Pauwels, 2006; Rasmussen, 1996; 1999). This orientation reproduces biopolitical dynamics: one imposed as governance over the virus, understood as self-contained object, and one regarding the struggle to cope with its “unstable connotations” (Esposito, 2008, p.14).

In contrast to bacteria, which can be measurable in micrometers, viruses range in size between 10 and 100 nanometers (Chandler & Roberson, 2009). The submicroscopic size of viruses delayed their visualization until the first half of the twentieth century. Their visual rendering was made possible with the invention of the electron microscope. In 1932, Max Knoll and Ernst Ruska, German physicists, successfully built a prototype transmission electron microscope (TEM), that subsequently was perfected by a team of scientists at the University of Toronto (Prebus, 1998; Rasmussen, 1996). The TEM constituted a major breakthrough in microscopy. Unlike light microscopes that operate by diffracting light and which have relatively limited resolve power, electron microscopes use beams of electrons whose wavelength is about 100,000 times shorter than visible light photons. This enables levels of magnification that are millions of times stronger than those obtained with light microscopes (Goldsmith & Miller, 2009). However, electron microscopes with such augmented capacity are neither user-friendly nor cost-effective instruments, and the operations required to capture the molecular content of viruses are labour intensive (Flegler, Jr., Heckman, & Klomparens, 1997; Villareal, 2005). They require a dedicated sealed environment and trained personnel to configure and calibrate the microscope and to operate its software (Goldsmith & Miller, 2009; University of Cambridge, 2012).

Prior to exposing a virus specimen to the electron beams of a TEM or a Scanning Electron Microscope (SEM), a multi-phase preparation is required. Biological specimens need to be prepared (i.e., protected and preserved) by means of fast freezing, using liquid nitrogen for the former or staining and coating the sample with the heavy metal osmium for the latter (Chandler & Roberson, 2009). Laboratory technicians supervise the process and prepare the specimen by applying a suitable chemical recipe comprised of exact amounts and proportions of chemicals and minerals so as to not compromise the sample’s integrity. Since these procedures are performed on very small-scale objects, they are prone to both errors and security and safety hazards (e.g., exposure to damaging agents) (La Berge, 1999). Thus, the success of this process depends on a variety of skills and calls for the intervention of a diverse range of well-trained professionals.

Once samples of viruses are selected, stained, or frozen, other procedures are conducted to further dissect and examine their anatomy. The visual product obtained from the electron microscope appears as a black and white image displaying a conglomerate of viruses (Figure 2). Its resolution varies according to the preparation technique, the goal established by the investigators, and the state of the microscope (Chandler & Roberson, 2009). This however, is by no means the final product. Colourful results are displayed in magazines and scientific journals (Chatterjee et al., 2013). The image of the H5N1 virus on the cover of New Scientist (Figure 1), for instance, could only be obtained after “a series of representations or renderings [was] produced, transferred, and modified as research proceed[ed] from initial observation to final publication” (Lynch, 1988, p. 202). These operations subject the initial microscopic scan to a series of reductive operations (Stengers, 2010) involving “selection” and “mathematization” (Lynch, 1988), which extract, isolate, dye, and re-colour the virus according to established conventions that break it down into its main components and simplify the appearance of its content, to facilitate analysis.

Throughout the process, consistency is sought in order to satisfy scientific mores and aesthetic forms that facilitate recognizing, demarking, and configuring viruses. Yet the nature of viruses and the numerous processes needed to obtain such visualizations invalidate such consistency. In other words, since the nature of viruses rejects universal definition and graphic homogenization, techno-aesthetic conventions intervene to ensure that the visualized object fits some minimum standard of identification. This intertwining of the normative and the singular within the process of visualization constitutes an encounter between the defining norms subtly imposed by biopower and the uncertainties, errors, and compromises characterizing the management of a substance that is anything but stable.

Instrumentalization or technical necessity?

In attempting to turn the “phenomena of study into manageable data” (Lynch, 1988, p. 204), visual processing strives to confirm basic visual references and expectations that facilitate recognition of the substance to be visualized. At the same time, these processes acknowledge the modes of representation established by the discipline. Manipulation of the image is fundamental in turning biological samples into readable images. According to Tufte (2001), operating various degrees of selection is a normative rule superseding all forms of visualizing and mapping. Refusing to do so would make the object too rich in detail to be effectively deciphered, thereby rendering the virus illegible. Like cartographers and map-makers, he explains, scientists have to make compromises and choose the amount and type of data to highlight. Despite yielding different results based on the microscopes, the methods and technologies employed, and the publics addressed, virus visualization undergoes increasing reduction that references previous representations or employs recurring conventional styles and layouts that dictate the way in which viruses—and submicroscopic substances in general—should be formally examined and visualized.

Subsequent processes of reduction “transform the symbolic into the geometric” (McCormick, 1988, p. 2), where the geometric reflects both a legibility requirement and a literal “geometrification” or mathematization of the sample (i.e., reduction of the sample to geometry). Lynch (1988) describes four phases in the reduction and transformation of an object of study into a diagram: filtering, uniforming, upgrading, and defining. In order to be properly examined, the H5N1 virus, for instance, had to be displayed according to typical layout patterns. The original black and white microscopic scans obtained from the TEM (Figure 2) were artificially coloured using a typical fluorescent green dye to make crucial components of the virus more discernible. Later, a single virus was selected, separated, and isolated from other viruses that may have appeared in the same scan. The latter operation was performed to eliminate “unused visibility” (Lynch, 1988, p. 209) or visual noise that may distract observation (filtering). The digitization process sharpened and redefined the image (see Figure 1) by assigning specific and solid colour fields (uniforming), stabilizing and consolidating their contours (upgrading), and locating the resulting illustration in a dark or empty background (defining). Having been dismembered, its molecular content separated, and its components meticulously defined to display proteins, lipids, RNAs, and other aspects of its structure, the virus displayed on the cover of New Scientist was presented as a singular artefact courtesy of a combination of design packages such as Chimera, Pymol, and Avogadro (“Avogadro-Molecular builder,” n.d.; “UCFC Chimera,” 2012; Schrödinger, 2013). The resulting visual product is an eidetic image destined to be only representative of a phenomenon, rather than reproducing the phenomenon in its entirety (Lynch 1988; 2006b).

Myers (2014) posits that “machinic analogies are not merely aesthetic flourishes of language or attractive figures of speech” (p. 159), but a metaphoric necessity. They can be seen as “‘lures’ that ‘vectorize’ practitioners’ imaginations and experimental inquiry” (Stengers, 2010, as cited in Myers, 2014). Visualization is “like the materialized refigurations that corporealize life in the form of information systems” (Myers, 2014, p. 162). In the visualization of submicroscopic substances that cannot be observed directly by the human eye (Mody, 2014; Ruivenkamp & Rip, 2014), incorporating tropes and cultural hints establishes a vivid comparison between familiar phenomena and the object displayed, helping both learned and partially educated audiences understand its significance. However, in the case of viruses, evoking cultural metaphors is unnecessary because viruses encapsulate assumed features and an affective gravity that is difficult to ignore. Both scientists and the general public are familiar with the rhetoric defining viruses as potential or looming pandemic threats (van Loon, 2002b).

Fluorescent and bright colours, odd intricacy, as well as the angular and well-defined shapes that visual representations of viruses display, seemingly confirm the rhetoric of fear. Colours and geometric shapes help scientists distinguish the constituent parts of viruses. Yet the propensity of most illustrations to display viruses in bold and fluorescent colours, as opposed to pale and light colours, communicates potential aggressiveness to viewers of these images. In the absence of an understanding of the visualization process, regular and geometric shapes become sources of suspicion. The variety and diversity of visualizations being circulated, combined with a general lack of visual literacy (Trumbo, 2006) and the popular narratives about viruses, exacerbate these suspicions and foster pervasive insecurity.

While the viruses portrayed in these images often appear to be artificially manipulated to look more threatening, they can also be understood as remnants of the technical processes employed in previous phases of visualization and as expressions of the aesthetic and formal obligations required to make virus identification possible. For instance, the green colour contained in Figure 1 was previously used in Figure 2 to highlight specific details in the virus requiring particular attention. Thus, its recurrence in Figure 1 can be justified as an expression of technical procedures, rather than as an instrumentalizing rhetorical stratagem. Likewise, the manipulation of shapes and patterns that makes viruses more comprehensible to the public often goes hand-in-hand with, and cannot be distinguished from, the ways in which they are perceived and feared. Thus, the very elements perpetuating typical perceptions of viruses as noxious substances also make them recognizable as viruses. In other words, in the visualization of viruses, necessary technical items are interwoven with the culture of fear that pervades them.

This coupling of the technical and rhetorico-cultural aspects of viruses reveals the extent to which biopower affects visualization. The very nature of viruses makes preparation and imaging processes heterogeneous, unpredictable, and barely manageable. In the case of H5N1, conventions and scientific mores (i.e., politics) imposed from above upon the object of study (i.e., life) had to be constantly adapted and modified in accordance with its behaviour, thus producing consistently different results. When analyzing the ways in which illustrations and renderings of viruses are manufactured, it is evident that assumptions and inherited cultural narratives and stereotypes affect the understanding of viruses and the reception of their visual representations, which, in turn, influences the conventions applied to make visualizations “familiar” to both scientific and general audiences. This tendency is a product of the cultural conceptualization of viruses and is an important element of biosociality.

Rabinow (1992) describes this phenomenon as an instance of biopower, and as a further development of a phenomenon by which “nature [is] modelled on culture understood as practice” (p. 241). This tendency, he argues, marks an era where “nature will be known and remade through technique and will finally become artificial, just as culture becomes natural” (p. 242). Pandemics and other widespread diseases, he continues, have already produced “the certain formation of new group and individual identities and practices arising out of these new truths” (p. 244) thanks to recurring outbreak narratives that collectively unify expectations about infectious phenomena and virus behaviour (see Wald, 2008). These narratives intertwine with, and partially determine, the technical construction and the aesthetics of visualization.

Elements of the visualization of viruses sit at the crossroad between technical needs and popular virus-oriented narratives, impacting laboratory practice. Assumptions, and with them attempts at instrumentalization, arise when the application of technical procedures matches visual expectations, triggering a quiet acceptance of these narratives. While it is virtually impossible to distinguish between the elements emanating from these narratives and those dictated by scientific and technical needs, they nonetheless constitute rules that enable the immediate identification of viruses while simultaneously arousing fear and conjecture.

If choice of colours and geometrical contours influences the ways in which we see and interpret virus illustrations, one cannot help but wonder what would happen to popular virus narratives, and the assumptions they evoke, if these elements were removed. Working in collaboration with virologist Dr. Andrew Davidson from the University of Bristol, glassblowing artist Luke Jerram tackled this question when he produced a series of giant glass molecular models of popular viruses (Boustead, 2009; Jerram, n.d.; Zielinska, 2009). Intrigued by the use of bright and arbitrary colouring in the molecular visualizations of viruses as portrayed in science and the media, he produced a series of transparent glass sculptures of the molecular structure of infamous viruses, including HIV, SARS, and H5N1 (see Figure 3). Despite its being an artificially added feature, many lay observers often assume that colouring in visualization reproduces the colours of viruses as they are found in nature (Jerram, n.d.). These individuals also tend to be easily persuaded by the rhetoric accompanying the use of these colours. However, members of the general public who attended Jerram’s sculpture exhibitions often found that their reaction to these viruses changed dramatically. Many remarked that their attention shifted from disgust and repulsion to genuine interest in the procedures that made the dissection of the viruses and the creation of the sculptures possible.

Figure 3


While Jerram’s project raises awareness of the arbitrary application of colours in the visualization of submicroscopic entities and its effects in the case of substances known for their negative connotations, it also underlines the extent to which such visualization practice reflects and, in turn, “adversely distorts the opinion of the viewer” (Boustead, 2009). By the same token, once the colours are gone, it becomes impossible for viewers to distinguish the many parts comprising the material structure of viruses. Artificial colouring may provoke adverse reactions, but it also is an indispensable tool in facilitating the work of scientists. Eliminating one of the major elements influencing cultural perceptions of viruses may partially neutralize one of the mechanisms controlling public reaction to the spread of viruses. Yet it also removes important information regarding specific viral substances. This underscores the interdependence of rhetoric and technical specificity in visualization.

The cultures of visualization: A race with no clear direction

The visualization industry comprises a variety of practices and projects that reproduce viruses in their external appearance, their molecular structure, and their behavioural patterns; that follow independent internal guidelines; and that respond to different research circumstances. Practices include the rendering of capsid and molecular models, 3D animations and simulations, visualizations from data, and illustrations for popular fruition. The recent proliferation of specialized software responds to requests made by scientific institutions wishing to communicate to larger audiences and to achieve enhanced visibility and prestige (Science Illustrated, 2011).

In the context of such diversity (and in spite of it), laboratories and graphic design studios appear to be competing with each other. This competition is not concerned with establishing a model for the industry or the discipline of visualization. Rather, it arises from a drive to simultaneously satisfy the mandate and research scopes of the laboratories served, and the audiences to which any given visualization is directed (Coopmans, 2014b; Myers, 2014). In other words, it is concerned with successfully managing and skilfully modulating, rather than taming and regulating, the “vital forces” (Thacker, 2009) that converge into visualization. Different parties involved in visualization processes seek to achieve several goals simultaneously: answering scientific questions, obtaining awards and research funding, and disseminating visual objects that attract the public with appealing aesthetics (Stafford, 1996; Structural Biology and Photosynthesis Research Group, 2014). Goals are reached by negotiating and discussing the information and aesthetics to be incorporated into the final visual object, and they involve members of laboratories and graphic design companies.

Seen from the perspective of science, this competition points inward, with each attempt at visualizing viruses aiming to solve an internal challenge. Each visualization gives its contribution to a precise study or problem which, in turn, requires a combination of differing techniques and aesthetics (Steinman & Steinman, 2007). Viewed from a popular science and/or popular culture viewpoint, the race is directed toward achieving public visibility and professional legitimacy. Here, visualization becomes a symbol of accountability, serving the goal of securing research benefits for laboratories and professional assignments for the graphic design studios (Structural Biology, 2014). Put simply, the products of scientific visualization fluctuate between using aesthetics that target and satisfy expectations of particular audiences, and facilitating message transmission. In both instances, the resulting visuals tend to be rather unique.

Awareness of the value of visualization in presenting scientific research to audiences—be it the general public or scientists—is increasing. The sophistication and accessibility of computer graphics, combined with the success of graphics and advertising industries, has made scientists more conscious of the immediacy and efficacy of images in reproducing and presenting their research (Steinman & Steinman, 2011). At the same time, using these tools to complement scientific practice has become a necessity. Releasing a simple electron microscopic scan processed through the built-in software no longer suffices. New funding policies and schemes, coupled with the privatization and fragmentation of such funds, make effective presentations and the popularization of scientific results an obligation. As a result, the proliferation of scientific visualization and illustrations has become common practice (Demeritt, 2000; Small & Mallon, 2007).

Judged as much for their scientific value as for their capacity to appeal to different audiences, virus images effectively become synecdoche for the work performed by entire laboratory teams. They become a flagship of productivity and accountability, increasing the scientist or the laboratory’s ability to lobby granting agencies for prestigious or more remunerative research funding (Jon Nield Group, 2010). It is not uncommon to enter a laboratory to find that its walls display award-winning covers and image awards (Steinman & Steinman, 2011). Laboratories and individual scientists compete to have their visualization images displayed on the covers of major journals (RIMAD, 2010) and participate in scientific visualization and micro-photography competitions, as exemplified by the multi-category competitions instituted by the U.K. based Wellcome Trust (Wellcome Images, 2011) and the U.S. National Science Foundation (Ceurstemont, 2011; Minogue, 2011; Nesbit & Norman, 2011).

The variety of products originating from laboratories and companies has accelerated the development of new tools and strategies to present and explicate submicroscopic data. However, this move also contributes to the molecularization of an industry whose diversification is encouraged by the competitiveness characterizing today’s new media and graphics industries. Each company uses different operating systems and software, catering to different individuals, engineers, microbiologists, and graphic designers. The New Scientist illustration of H5N1 and Jerram’s reproduction of the same virus are illustrative of the variety characterizing the visual production of viruses. Both employ similar scientific data for their models, and both have appeared in popular science magazines. However, their different objectives, media, and materials have caused the two models to differ radically.


The diverse range of virus visualizations produced with a continuously expanding number of customizable software options reflects the “evolving collection of new technologies that facilitate the creation of captivating images” (Trumbo, 2000). Endeavouring to combine information and aesthetics to satisfy the variety of goals supported by laboratories and editorial teams, scientific visualization unavoidably disseminates heterogeneous products. This trend satisfies neither the desire to conquer the furtiveness of viruses, nor the desire to produce standardized criteria for the consistent representations of viruses.

Visualization struggles to control the virtual object called “virus.” The variety characterizing this field, and the nature of viruses themselves, does not allow this to happen. Yet attempts at managing these substances can be found at all stages leading to, and included in, the visual representation of viruses. Recurrent elements ensure that images of viruses remain anchored to visual layouts that both picture them in isolation and that respond to visual tropes and popular narratives. At the beginning of his Picture Control, Rasmussen (1999) defines the nature of scientific change as an oxymoronic business, noting that “science must conserve and accumulate, yet science must continually progress and thus overthrow its own past. Science grows by radical conservation, or perhaps by conservative revolution” (p. 3). A similar set of dynamics can be recognized in the constant transformation of visualization. While here the term “conservative” reproduces certain practices specific to the laboratory environment in order to conform to established knowledge, it does not credit the role of the viewer in perpetuating this knowledge. Neither does it reflect the “encounter” (Esposito, 2008) between the vitality of a rebellious object of study (in this case, the virus) and the scientific, social, and political desire to order and constrain such objects (that is, the management over the virus and of the virus).

In general, the practice of visualizing viruses appears to chase an object that refuses to be packaged in a definite form. Seen from a different perspective, visualization is a major contributor to the flourishing and rapid diversification of software development and visual innovation. It is thanks to, and in the tensions created between, the persistence of default patterns for the formal representation of viruses, and the creative modulation of their visual appearance, that scientific visualization manages to maintain its heterogeneity.

Read as an encounter of bíos and politics, regulation and recommendation, the variety characterizing visualization and transformation stems from the indeterminacy of viruses. It reflects a desire to tackle such indeterminacy and to defeat, if only gradually, the visual, technical, and cultural rules that dominate the current visual appearance of viruses. In fact, this diversity manifests by means of small but ceaseless changes and adjustments that function as major incentives fuelling the renewal of the field of scientific visualization itself.


Technical instructions were contributed by: David Sanchez-Tatay, Wellcome Trust Stem Cell Research Lab, University of Cambridge, UK; Dolores and David Steinman, Biomedical Simulation Lab (University of Toronto); Jon Nield from the Structural Biology and Photosynthesis Research Group (Queen Mary, University of London), and the RIMAD (Research Institute for Media, Art and Design) talk series (University of Bedfordshire). This research was made possible thanks to a British Academy Visiting Scholar Fellowship in Cambridge, UK, under the supervision of Dr. Jussi Parikka.


Avogadro- Molecular Builder. (2013). Avogadro. URL: [October 15, 2009].

Barad, Karen. (2007). Meeting the universe halfway: Quantum physics and the entanglement of matter and Meaning. Duke University Press.

Boustead, Greg. (2009, October 15). At the edge of perception. Luke Jerram’s work explores the limits of science and art, challenging the boundaries of both. Seed Magazine. URL: [July19, 2011].

Bozzola, John J., & Russell, Lonnie Dee. (1999). Electron microscopy : Principles and techniques for biologists (p. 670). Sudbury, Mass.: Jones and Bartlett.

Ceurstemont, Sandrine. (2011, February 17). New Scientist TV: HIV as you’ve never seen it before. New Scientist. URL: [October 26, 2011].

Chandler, Douglas E., & Roberson, Robert W. (2009). Bioimaging : Current concepts in light and electron microscopy (p. 440). Sudbury, Mass.: Jones and Bartlett Publishers.

Chatterjee, Somnath, Roy, Anirban, Laskar, Aparna, & Swarnakar, Snehasikta. (2013). Electron microscopy in the perspective of modern biology: Ultravision and ultradimension. In A. Méndez-Vilas (Ed.), Current microscopy contributions to advances in science and technology (pp. 891–902). Badajoz, Spain: Formatex Research Center.

Coopmans, Catelijne (2014a). Introduction: Representation in scientific practice revisited. In J. Vertesi, M. E. Lynch, S. Woolgar, & C. Coopmans (Eds.), Representation in scientific practice revisited. Cambridge MA: The MIT Press.

Coopmans, Catelijne. (2014b). Visual analytics as artful revelation. In J. Vertesi, M.E. Lynch, S. Woolgar, & C. Coopmans (Eds.), Representation in scientific practice revisited (pp. 37–60). Cambridge MA: The MIT Press.

Demeritt, David. (2000). The New Social Contract for Science: Accountability, Relevance, and Value in US and UK Science and Research Policy. Antipode, 32(3), 308–329.

Esposito, Roberto. (2008). Bíos: biopolitics and philosophy. University of Minnesota Press.

Flegler, Stanley L., Heckman, John W. Jr. & Klomparens, Karen L. (1997). Scanning and transmission electron microscopy: An introduction (New Ed.). OUP USA.

Foucault, Michel. (2003). Society must be defended : Lectures at the Collège de France, 1975-76. (M. Bertani, A. Fontana, F. Ewald, & D. Macey, Eds.). New York: Picador.

Foucault, Michel. (2007). Security, territory and population (1st ed.). Palgrave.

Goldsmith, Cynthia S., & Miller, Sara E. (2009). Modern uses of electron microscopy for detection of viruses. Clinical Microbiology Reviews, 22(4), 552–563.

Haraway, Donna J. (1997). Modest_witness@second_millennium.FemaleMan©_meets_OncoMousetm:Feminism and technoscience. New York: Routledge. New York: Routledge.

Jerram, Luke. (n.d.). Personal Website. URL [December 4, 2011].

La Berge, Ann Elizabeh Fowler. (1999). The history of science and the history of microscopy. Perspectives on Science, 7(1), 111–142.

Latour, Bruno. (1988). Science in action. Open University Press.

Latour, Bruno. (1996). On actor-network theory: A few clarifications. Soziale Welt, 47(4), 369–81.

Latour, Bruno, & Woolgar, Steve. (1979). Laboratory life: The social construction of scientific facts. Sage.

Lynch, Michael. (1988). The externalized retina: Selection and mathematization in the visual documentation of objects in the life sciences. Human Studies, 11(2-3), 201–234.

Lynch, Michael. (2006a). Discipline and the material form of images: An analysis of scientific visibility. In L. Pauwels (Ed.), Visual cultures of science: Rethinking representational practices in knowledge building and science communication (pp. 195–221). Hanover, NH: Dartmouth College Press.

Lynch, Michael. (2006b). The production of scientific images: Vision and re-vision in the history, philosophy, and sociology of science. In L. Pauwels (Ed.), Visual cultures of science : Rethinking representational practices in knowledge building and science communication (pp. 26–40). Hanover, NH: Dartmouth College Press.

Macey, David. (2009). Rethinking biopolitics, race and power in the wake of Foucault. Theory, Culture & Society, 26(6), 186–205.

McCormick, B. H. 1988. Visualization in scientific computing. ACM SIGBIO Newsletter 10(1), 15.

Minogue, Kristen. (2011). Visualization challenge illustration. Science 331(6019), 848–849.

Mody, Cyrus C.M. (2014). Essential tensions and representational strategies. In C. Coopmans, J. Vertesi, M.E. Lynch, & S. Woolgar. (Eds.), Representation in scientific practice revisited (pp. 223–248). Cambridge MA: The MIT Press.

Myers, Natasha. (2014). Rendering machinic life. In C. Coopmans, J. Vertesi, M.E. Lynch, Michael & Woolgar, Steve (Eds.), Representation in scientific practice revisited (pp. 153–176). Cambridge MA: The MIT Press.

Nesbit, Jeff, & Norman, Colin. (2011). 2010 visualization challenge. Science, 331(6019), 847.

Pauwels, Luc. (2006). A theoretical framework for assessing visual representational practices in knowledge building and science communications. In L. Pauwels (Ed.), Visual cultures of science: Rethinking representational practices in knowledge building and science communication (pp. 1–25). Hanover, NH: Dartmouth College Press.

PDB. (n.d.). RCSB PDB Protein Data Bank. URL: [November 25, 2012].

Prebus, Albert F. (1998). Events at the University of Alberta and the University of Toronto, leading to the first North American electron microscope. Micron, 29(4), 289–292.

Rabinow, Paul. (1992). Artificiality and enlightment: From sociobiology to biosociality. In M. Biagioli (Ed.), The science studies reader (pp. 407–416). New York: Routledge.

Rasmussen, Nicolas. (1996). Making a machine instrumental: RCA and the wartime origins of biological electron microscopy in America, 1940-1945. Studies in History and Philosophy of Science Part A, 27(3), 311–349.

Rasmussen, Nicolas (1999). Picture control: The electron microscope and the transformation of biology in America, 1940–1960. Stanford University Press.

RIMAD (Research Institute for Media, Art and Design). (2010). Open repository. University of Bedfordshire. URL:[June 6 2013].

Ruivenkamp, Martin, & Rip, Arie. (2014). Nanoimages as hybrid monsters. In J. Vertesi, M.E. Lynch, S. Woolgar, & C. Coopmans (Eds.), Representation in scientific practice revisited. Cambridge MA: The MIT Press.

Science Illustrated. (2011, February 22–23). A symposium on visualizing science. McLeod Auditorium, Medical Science Building, University of Toronto.

Pymol. (2013). A molecular visualization system on an open source foundation, maintained and distributed by Schrödinger. URL: [October 17, 2013].

Small, Bruce, & Mallon, Mary (2007). Science, society, ethics, and trust: Scientists’ reflections on the commercialization and democratization of science. International Studies of Management and Organization, 37(1), 103–124.

Stafford, Barbara Maria. (1996). Good looking: Essays on the virtue of images. Cambridge, MA: The MIT Press.

Steinman, David, & Steinman, Dolores. (2007). The art and science of visualizing simulated blood-flow dynamics. Leonardo, 40(1), 71–76.

Steinman, David, & Steinman, Dolores. (2011). Toward new conventions for visualizing blood flow in the era of fascination with visibility and imagery. In O. Grau & T. Veigl (Eds.), Imagery in the 21st century (pp. 129–148). Cambridge MA: MIT Press.

Stengers, Isabelle. (2010). Cosmopolitics I. Minneapolis: University of Minnesota Press.

Structural Biology and Photosynthesis Research Group. (2012). Research groups. Queen Mary Univerity of London. URL: [June, 4, 2014].

Thacker, Eugene. (2009). The shadows of atheology: Epidemics, power and life after Foucault. Theory, Culture and Society, 26(6), 134–152.

The Jon Nield Group in Biochemistry and Structural Biology. (2010). Structural biology and photosynthesis. London, UK: Queens Mary, University of London. URL [November 4, 2014].

Trumbo, Jean. (2000). Seeing science. Science Communication, 21(4), 379–391.

Trumbo, Jean. (2000). Seeing science. Science Communication, 21(4), 379–391.

Trumbo, Jean. (2006). Making science visible: Visual literacy in science communication. In L. Pauwels (Ed.), Visual cultures of science: Rethinking representational practices in knowledge building and science communication (pp. 226–284). Hanover, NH: Dartmouth College Press.

Tufte, Edward R. (2001). The visual display of quantitative information. Cheshire, CT.: Graphics Press.

UCFC Chimera. (2012). UCFC chimera. URL [October 12, 2013].

University of Cambridge. (2012). DoITPoMS, dissemination of IT for the promotion of materials science. [Online instructional manual] URL: [October 18, 2010].

van Dijck, José. (2005). The transparent body: A cultural analysis of medical imaging. University of Washington Press.

van Loon, Joost. (2002a). A contagious living fluid: Objectification and assemblage in the history of virology. Theory, Culture and Society, 19(5/6), 107–124.

van Loon, Joost (2002b). Risk and technological culture: Towards a sociology of virulence. New York, NY: Routledge.

van Regenmortel, Mark H.V., & Mahy, Brian W.J. (2004). Emerging issues in virus taxonomy. Emerging Infectious Diseases, 10(1), 8–13.

Villareal, Luis P. (2005). Viruses and the evolution of life. Washington DC: ASM Press.

Wald, Priscilla. (2008). Contagious: Cultures, carriers, and the outbreak narrative (1st ed). Duke University Press Books.

Wellcome Images. (2011). Wellcome images awards. URL: [Oct 12, 2013].

Zielinska, Edyta. (2009, October 2). Fragile flu, siliciferous smallox. The Scientist. URL:—siliciferous-smallpox [November 25, 2010].


  •  Announcements
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Current Issue
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Thesis Abstracts
    Atom logo
    RSS2 logo
    RSS1 logo

We wish to acknowledge the financial support of the Social Sciences and Humanities Research Council for their financial support through theAid to Scholarly Journals Program.