Silhouettes analysis: a posthuman method for visualizing and examining the material world

Introduction

In recent decades, postmodern, poststructuralist, and social constructivist theories, and the methodologies they have informed, have been criticized for focusing primarily on human actors, human discourses, and human actions. In response, a range of concepts has been developed in the social sciences, gender studies, cultural studies, humanities, and science and technology studies to consider not only human but also nonhuman actors such as artifacts, technologies, and animals as constitutive elements in the production of the social world. These concepts reject an unquestioning use of the human/nonhuman dualism, emphasizing instead the importance and diversity of the material world, which is constituted in the entanglements of material actors (e.g. Barad, 2003; Braidotti, 2014; Haraway, 1991, 2007; Latour, 2005). A key concern for current qualitative inquiry in various research areas is finding ways to account for these entanglements—that is, to develop methods that allow researchers to criticize human-centered analysis in order to show that the dualism of human/nonhuman is not always empirically valid. Schadler (2014, 2019), for example, suggests a new-materialist ethnography for family research; Kim McLeod (2014) decenters the human during different stages of research; I analyze verbal und nonverbal communication and references to things that were present or absent during interviews with the elderly in order to retrace continuities and discontinuities of embodiment processes (Höppner, 2015a, 2015b, 2017); and Feely (2019) develops a new-materialist assemblage analysis for analyzing narrative data in the field of disability studies. ¹ The posthuman methods thus far developed aim to question and decentralize the human being as the sole actor in the creation of family, research processes, and disability; for empirical material, interview transcripts have been the prime sources—that is, verbal and partly nonverbal data that were written down. This paper instead explores what we learn about the material world when we use and examine image details, which are made into silhouettes in several steps—that is, black formations on a white background—as empirical material for posthuman analysis. It will be shown that silhouettes analysis is a good complement to verbal methods in the effort to better understand the material world and contribute to the critical questioning of a human-centered qualitative form of research.

In Karen Barad’s (2003, 2007) agential realism, silhouettes show spatially limited forms that are inseparably connected to their surroundings, but are separated by agential cuts from these surroundings for analytical purposes. A silhouettes analysis is an image-analytical method that contributes to a posthuman qualitative research. Instead of taking for granted and reproducing the dualism of human/nonhuman by examining predetermined entities such as a person or object, a silhouettes analysis visualizes and examines the formations that have moved at the moment when a picture is taken.

A silhouettes analysis makes it easier to ask more generally which formations constitute and characterize a social phenomenon. In order to illustrate this argument, I will describe a silhouettes analysis using the phenomenon of old age as an example. This phenomenon is particularly well suited to this purpose, because age is generally considered to be located in the human body, whereas a posthuman interpretation of empirical data shows something different. In an ethnographic research project in several nursing homes in western Germany, I interviewed social workers, observed their group activities with residents, took photos of the residents, and analyzed pictures that are published on websites of nursing homes. It turned out that a reduction of old age to the human body does not go far enough, because every-day and commonplace objects such as clothes and jewelry, and technical artifacts such as walking frames and hearing aids might also characterize the materiality of old age.

Recently, the idea of human decentralization has found attention in gerontological theories. Using the term ‘material gerontology’ (Höppner and Urban, 2018), gerontological scholars point to both their skepticism toward the idea that ageing processes take place solely in humans as isolated entities and their openness to considering nonhuman kinds of materiality as being involved in ageing processes. In doing so, neither human nor nonhuman actors are to be understood as natural phenomena. In order to initiate and foster a debate on human and nonhuman actors as being essential and equal components in the production of age and ageing, scholars of material gerontology deal with various kinds of materiality from a theoretical perspective and thus make possible new perspectives on our understanding of old age. They conceptualize various interrelations of age and materiality, such as age and body (Höppner, 2015a, 2015b, 2017; Höppner and Urban, 2018), age and artifacts of everyday life and technology (Artner et al., 2017; Depner, 2015; Endter, 2016; Wanka and Gallistl, 2018), and age and spaces (Hahmann, 2018; Wanka and Oswald, 2020).

However, qualitative methodologies and methods used to investigate age and ageing empirically are still often focused solely on human beings—that is, on human communication, generated through interviews (Amrhein, 2008; Hartmann-Tews et al., 2008; Lundgren, 2013) and conversational analysis (Rossow and Koll-Stobbe, 2015; Thimm, 2000), on human thoughts and feelings, generated by autoethnographic research (Melina, 2008), and on images of people in later life (Martin, 2012; Richards et al., 2012). Methodologies and methods that question how the material diversity of old age can be generated and analyzed are rare in gerontological research (Höppner, 2015a, 2015b; Urban, 2017)—a gap that this contribution attempts to fill.

In order to introduce the method of silhouettes analysis, I sketch out the foundations of a posthuman methodology, which is informed by Barad’s (2003, 2007) agential realism and link them to ideas of image analysis. An exemplary silhouettes analysis makes two things clear. First, the critical questioning of a human-centered qualitative form of research helps in the effort to gain new knowledge concerning the features of the materialities of old age. Second, a silhouettes analysis makes it possible to disturb the ‘practiced’ researcher’s view and to question the apparent lack of ambiguity. In this respect, silhouettes analysis is also a research tool stimulating reflection, a tool that can both inspire their own interpretation of data and make them more aware of their own assumptions. I end with a discussion of the advantages and limitations of silhouettes for gerontological and posthuman analyses.

Posthuman methodology: foundations

There are different types of posthuman. In my understanding, posthuman does not mean anti-human or exclusively human enhancements, nor does it mean something coming after humanism. For me, posthuman means the openness to and consideration of the multifaceted, diverse, and complex world in which human and nonhuman actors are linked to and shaped by each other (Braidotti, 2014; Ulmer, 2017). In this understanding of posthuman, I see many points of reference to Barad's work on agential realism (Barad, 2003, 2007). I use these to develop a posthuman methodology for silhouettes analysis.

Based on the epistemological framework of quantum physics, Barad (2003) has developed an agential realist account of ethics, ontology, and epistemology. Barad (2003: 815) argues that ‘phenomena are the ontological inseparability of agentially intra-acting “components”’—that is, phenomena do not have any pre-existing relata such as a determinist nature. Rather, phenomena are constituted by the drawing of boundaries. Barad (2003: 822, 2007: 208) names these practices intra-actions as opposed to interactions, which presume the prior existence of independent entities and relata. Intra-actions are material-discursive practices through which the relation between materiality and meaning emerges. The term ‘material-discursive practices’ points to Barad’s (2003: 818) assumption that the discursive is always material and the material always discursive. Intra-actions are ongoing boundary-drawing practices that constitute components of a phenomenon—in a later work, Barad (2007) calls them ‘matterings’—that is, temporary, physical existences with altering boundaries, provided with specific properties. Hence, such matterings do not have a solid or fixed essence; there is not the body or the thing in itself. Rather, there is something Barad calls the ‘material dimensions of agency’ (Barad, 2007: 194), where agency is derived from the intra-actions of matter and meaning.

For her research purposes Barad argues that matterings are always embedded in an ‘apparatus of observation’ (Barad, 2003: 815)—that is, a specific entanglement of intra-actions. She admits of no classical ontological condition of exteriority between observer and observed; rather both observer and observed are enacted through intra-actions, both inseparably belong to the apparatus of observation. Which matterings count as matterings of a phenomenon depends on the agential cuts that are enacted during the research process. Barad (2003: 815) uses the term of ‘agential cut’ in contrast to the Cartesian cut, which she defines as an inherent distinction between subject and object. Agential cuts enact ‘a local resolution within the phenomenon of the inherent ontological indeterminacy’ and thus effect a form of relata that was produced only through ‘agential separability—the local condition of exteriority-within-phenomena’ (Barad, 2003: 815).

Barad’s ideas help to clarify the analytical focus of a material gerontology: Not a fixed entity that is taken for granted and called ‘old person’ or ‘old body’ is the topic of an analysis but the material-discursive phenomenon that is called ‘old age’. This phenomenon can be characterized through ongoing and specific boundary-drawing practices that constitute specific matterings. It is important to say that these matterings are radically embedded in an environment and are only separated from the environment through agential cuts.

For posthuman qualitative research on old age, it is important to (1) visualize the matterings that are constituted through boundary drawing practices, and (2) analyze the properties with which those matterings are provided. These two steps help us to understand which kinds of matterings characterize the phenomenon of old age.

A silhouettes analysis is helpful here because silhouettes present specific matterings. They show spatially limited formations that are inseparably connected to their surroundings; only through agential cuts can silhouettes be separated from their surroundings for analytical purposes—that is, silhouettes have to be contextualized at the place where a picture was taken and thus in the research apparatus in which a silhouette is generated, visualized, and analyzed. The boundaries of a silhouette make visible the agential cuts that were enacted. While gerontological researchers most often choose Cartesian cuts that separate human from nonhuman, with the result that they have to choose where to locate age (we know that most often the human body is chosen), agential cuts enable us to think of old age as a material-discursive phenomenon that is constituted through the drawing of boundaries. A silhouettes analysis helps to visualize and characterize these matterings.

Barad’s idea concerning agential cuts also enables reflection on the research process and the material-discursive practices that influenced it. Since agential cuts are always practices that include matterings in a phenomenon while excluding other possible matterings (Barad, 2003: 824), the following questions arise: Which possible matterings of an analyzed phenomenon are visualized through silhouettes and which matterings are not? How are these visualizations presented—as a single snapshot or as a sequence of silhouettes that try to trace ongoing intra-actions and thus a process of becoming? Based on which patterns are boundaries drawn and what forms of inclusion and exclusion are thereby produced and stabilized?

Why silhouettes?

I had arrived a few minutes too early at the nursing home where I had an appointment for an interview. In the big sun-flooded entrance hall, I sat down and waited for my appointment. I observed a lively bustle there. I was irritated to find that I couldn't immediately see what was moving back and forth in front of me because of the sun shining in my face—I saw silhouettes of surprisingly diverse and unusual shapes. At second glance, I recognized residents of the nursing home who were passing me in rows with their walking frames, residents with hands outstretched pushing drip stands, residents in wheelchairs. These observations and what I perceived to be a disruption of my viewing habits led me to develop silhouette analysis. As a material gerontologist researcher, encouraged by Pink (2001: 4), who states that researchers should create new methodologies and methods rather than attaching ‘the visual to existing methodological principles and analytical frames’, I became fascinated by the idea of making old age visible through silhouettes. From the field of performance art, we know the value of using silhouettes as concrete tools. However, image analysis using silhouettes is unusual, although I would like to emphasize its advantages.

The alienation effect provoked by a silhouette makes it possible to see and perceive the source of the silhouette in a different way. A standard image analysis of photographs usually makes use of Cartesian cuts, because it differentiates between entities such as human and walking frame (see e.g. Breckner, 2007). ² This type of distinction is based on, among other things, different color schemes, textures of materiality, designs, shapes, and also on our awareness of what a person or a walking frame looks like. This recognition effect is disturbed by silhouettes. They make it difficult to see previously defined entities because color schemes, textures of materiality, designs, and often even familiar shapes are eliminated. Latour’s (2000) idea of symmetry is taken literally here, conceiving a flat ontology where all elements, including nonhuman elements, are given equal status (McLeod, 2014). Black silhouettes on a white background do not show differentiations that allow the viewer to come to a conclusion regarding the hierarchy of the viewed object in terms of quality or to distinguish between human and nonhuman. Since silhouettes are enacted by agential cuts in a Baradian sense, they make visible matterings that might not fit into familiar categories, such as human or nonhuman. This ambiguity is intentional, because it serves to problematize more generally what could be seen, described, and analyzed as a formation that constitutes a social phenomenon. In this way silhouettes analysis helps to prevent the dualism of human/nonhuman from being reproduced without reflection. However, the gain in knowledge not only relates to the data to be analyzed, but also to the researchers themselves: The alienation effect that a silhouette triggers could be used to irritate one's own view as a researcher, with the aim of critically questioning one's own expectations and assumptions. Such a level of reflection is not provided by standard image analysis.

Producing silhouettes

Between the time of the first observation of silhouettes in the nursing home and the time when the silhouettes are prepared for analysis, various steps are necessary. In a Baradian sense, it is important to emphasize that these steps are to be understood as material-discursive practices that constitute a specific apparatus of observation. It is only in this apparatus that observer and observed as research participants, a camera as research equipment, and the nursing home as the context of the photographs are enacted through intra-actions; these differentiations make it possible to recognize and address them as such. And it is only through this apparatus that agential cuts are enacted that produce the formations that are defined as matterings of a specific phenomenon—in this research example, the phenomenon of old age.

It is important to ask on what principle agential cuts are performed—thus enabling them to distinguish matterings from their surroundings—if not on the principle of presupposed ontological units like the human body or distinctions like those between human and nonhuman, as in the case of Cartesian cuts. For a silhouettes analysis, agential cuts are enacted on the principle of movement—that is, matterings are defined as such because they moved during the moment when a photograph was taken while the surroundings of the moving formation were static. Silhouettes are thus differentiated on the basis of the distinction between moving and not-moving. Here Barad’s point that the concept of agential cuts does not assume that dualisms are dissolved is important, since continual reconfigurings of the natural-cultural world produce ongoing differentiations (cf. Barad, 2007). However, Barad feels the need to reflect on the principles by which agential cuts are performed and whether it is possible to gain new knowledge if distinctions other than the standard ones are used.

Following the example provided by Pauwels’ (2013) dual typology, the use of both researcher-produced visuals and visual materials from websites can produce silhouettes as long as they satisfy the principle of showing moving formations. On the basis of these visuals, it is possible to mark the formations as silhouettes. It is easiest if the visuals show a silhouette. In order to achieve the alienation effect, colors and textures must be removed. Only then can a local resolution within the phenomenon of the inherent ontological indeterminacy be achieved; then a silhouette presents a local condition of exteriority-within-phenomena (cf. Barad, 2003: 815) and thus a form of relata that was enacted only through agential separability within an apparatus of observation. Since photos showing moving formations do not always meet this condition, black coloring can be used to emphasize the formations. This can be done with a PC program or a black marker. The alienation effect of the silhouette is greatest when the background is undifferentiated. If a black marker is used, the silhouette can be cut out and pasted onto a white background.

The steps described make it clear that a silhouette is never the moving formation. Rather, silhouettes are matterings that were constituted through intra-actions, that is, several boundary-drawing practices that are both material—involving different materials and physical activities—and discursive in the sense that certain assumptions such as the distinctions between moving and not-moving are made and certain theoretical assumptions are used, such as Barad's agential realism in this research.

Analyzing silhouettes

A silhouettes analysis is carried out in two steps. In a first step, the researcher describes the shape of a silhouette to better understand the matterings that constitute a phenomenon. This step is helpful because silhouettes do not point up known formations of a phenomenon by means of Cartesian cuts; rather they make unusual formations visible by means of agential cuts. To describe the shape of a silhouette the use of optical elements of image analysis are helpful, such as sizes, lines, areas, and proportions (cf. Prosser, 2011). In a second step the researcher draws conclusions from the descriptive analysis—that is, the researcher examines the properties with which these matterings are provided. Both steps are presented in the following, using first a single silhouette and then a sequence of silhouettes.

First Step: describing silhouettes

Describing a single silhouette

The first silhouette is exemplary of many other similar formations that were observed in several nursing homes during the research project. In terms of area and outline, the silhouette has nearly the shape of an isosceles triangle. The silhouette has an ‘s’-line on the left and the right leg of the triangle—the boundaries of the silhouette on the left and the right side are similar in their forms— whereas the leg on the right side is less clear; a kind of curved bridge links a compact shape with a half circle at the top to a more interrupted shape with a half circle on the ground (Figure 1).

OPEN IN VIEWER

Figure 1.

Area and outline.

The inner area of the silhouette is interrupted by several blank spaces; the black and white areas are arranged irregularly.

Two contact points seem to connect the left and the right part of the silhouette (Figure 2). Especially the upper contact point appears comparatively thin, though it has the task of balancing the weight of the formation and coordinating its movement.

OPEN IN VIEWER

Figure 2.

Contact points I.

The silhouette contacts the ground at two points; the third possible contact point does not touch the ground but rather hovers in the air (Figure 3). The middle contact point is a rectangle that seems to give the silhouette stability and secures its equilibrium. However, since the contact point constantly changes due to movement (partly rolling), it is more fragile than it seems at first sight.

OPEN IN VIEWER

Figure 3.

Contact points II.

The view of the contact points also shows that the larger part of the silhouette is situated at the bottom. It seems that this arrangement of shape allows or facilitates the movement model for a while.

Describing a sequence of silhouettes

The second silhouette was selected for maximum contrast; only a few of these formations were observed during the research project. The silhouette has the shape of an irregular rectangle—that is, none of the legs of the rectangle resembles the others, and each leg on its own is irregular in shape. The inside area of this silhouette is interrupted by more blank spaces than there are in the first silhouette. There is only one contact point that seems to connect the left and the right part of the silhouette and two relatively distant contact points where the silhouette touches the ground; these contact points are small relative to the size of the silhouette (Figure 4, right side).

OPEN IN VIEWER

Figure 4.

Movement model.

An additional silhouette presents a movement model (Figure 4, left side). While one part of the silhouette remains constant as a component, the other part is replaced. The change from one temporary part to another temporary part requires the constant part of the silhouette to change in shape: While it bends down to the first temporary component, it appears more stretched and straightened in the contact with the second temporary component—even though the contact points are at similar heights.

Advantages of a silhouettes analysis for gerontological research

The silhouettes analysis demonstrated two things: first, it generated insights into features of the matterings of the phenomenon of old age that a classical image analysis would not have made possible. And secondly, it produced an alienation effect that disturbed typical viewing habits and their associated presumptions. Both of these factors can stimulate gerontological research.

Not only in gerontological research are physical characteristics such as grey or white hair, wrinkles and a stooped posture considered signs of age. The problem with these features is their inherent differentiating logic and their normative charge: either a person has gray or white hair, wrinkles and a stooped posture and is therefore considered old, or s/he does not have them and is therefore not considered old—both ascriptions are linked to specific behavioral expectations. Accordingly, gerontological research has for some years also been concerned with the question of what physical behavior older people practice and whether this behavior can contribute to anti-ageing (e.g. Höppner, 2011; Pfaller, 2016; Schroeter, 2012). Even if this research has generated important findings, it remains rooted in the logic of old/not old or ageing/anti-ageing. The silhouettes analysis does not base itself on these differentiations, but rather on the basis moving and not-moving. This allows other characteristics of old age to be generated: material diversity, formations consisting of two parts, no permanency, no constancy, and a certain instability are features that relate to the shape of the materialities of old age—characteristics that are not as common in gerontological research as are physical characteristics and abilities. These features offer the opportunity to show that old age is not necessarily to be characterized according to a binary logic; in an agential realist reading, old age can also be conceptualized as a continuum of materializations. Here it becomes clear how important it is to reflect on the assumptions and principles that a given form of research directs, and—closely related to this—whether Cartesian cuts or agential cuts are used because they determine what can be generated, seen, and analyzed.

Closely related to this and within the framework of material gerontology, the silhouettes analysis makes it possible to visualize material formations of old age and gain empirical findings that provide an understanding of old age that is not human-centered. In this respect, these findings support the idea that age should be understood as a distributed age in which human and nonhuman materialities, such as persons, everyday objects, technical artefacts, and spaces are constituted and at the same time produce what we call ‘old age’ (Höppner, 2021). These findings also touch a central question of material gerontology, dealing with a rethinking of the ontology of age and thus with the question of what actually is an old person: Where does s/he start, where does s/he end? And aren't the boundaries of a person only brought about as such by intra-actions (cf. Höppner and Urban, 2018)? Silhouettes analysis supports an empirical investigation of these questions of material gerontology since it alienates well-known shapes of old age.

Gerontological researchers who want to do posthuman analysis are not protected from falling back into familiar thought patterns. Images that are received and internalized as ‘normal’—that is, most often a straight human with a clearly definable shape having two arms and two legs, a human that has corporeal abilities and thus a body that does not have to be supported by medical artefacts—are often very powerful. Alternative body formations run the risk of being constructed as deviations from this norm. The silhouettes analysis provides gerontological researchers with room for reflection, because it makes it clear that such a ‘normal’ body shape is empirically unlikely in old age. In this respect, silhouettes not only make it possible to represent, but also to see and perceive old age differently (cf. Traue et al., 2018).

Although silhouettes analysis is a posthuman method, it enables us to learn more about the category of human. Using old age as an example, the analysis demonstrated that human beings are to be understood in relation to other components and that they are hardly in a position to shape their own movements completely through their actions. Movement seems rather to be constituted by the joint actions of different components.

Finally, silhouettes analysis can contribute to a better understanding of the process in which ageing takes place. The fact that old age can be produced interactively has already been conceptualized (Laz, 1998; Schroeter, 2012) and empirically reconstructed (e.g. Lundgren, 2013; Rossow and Koll-Stobbe, 2015; Thimm, 2000). However, the focus of the empirical research was on written data from interviews. In keeping with Barad’s idea of material-discursive practices, an analysis of material formations through silhouettes complements empirical research on old age.

Advantages and limitations of a silhouettes analysis for posthuman research

Posthuman theories aim to decentralize the human, reject an unquestioned use of the dualism of human/nonhuman, and emphasize the importance of the material world in the production of the social. This article has introduced the method of silhouettes in order to show what we learn when we do not use common qualitative methods and written data but instead image details of moveable matterings as in Barad’s agential realism in order to contribute to the critique of human-centered analysis.

One advantage of using silhouettes is the gaining of new insights into a phenomenon because silhouettes encourage researchers to ask for the formations that characterize a phenomenon. In this paper, the agential cuts made in the silhouettes analysis along the boundary between moving and not-moving generated features of the matterings of the phenomenon of old age, and particularly of the movements or movement models. However, researchers who plan to use silhouettes as a method should be aware that in their analyses they will necessarily draw boundaries, and will thus be able to examine only specific matterings of a phenomenon.

Another advantage of silhouettes is that they disturb one’s own viewing habits. Due to the alienation effect that they generate, silhouettes help to prevent researchers from seeing something whose representation they are already familiar with. Taking old age as an example, it can be said that the silhouettes that are examined in this paper certainly do not correspond to what we commonly think of as old age. The alienation effect also allows researchers to initially leave out the context of a silhouette and concentrate solely on the silhouette. Since silhouettes are inseparably connected to their surroundings, but are separated by agential cuts from these surroundings for analytical purposes, a second analytical step contextualizes the silhouettes within an environment and frames them as formations of a specific phenomenon.

From my own experience as a researcher, I can say that an agential realistic analysis of interview material is not always easy (Höppner, 2015b). Language is expressed by people and is based on human patterns of thinking: words are determinate entities that are associated with meanings, and grammatical structures presuppose relations as given. Both complicate a posthuman analysis because they make it difficult to think beyond the human/nonhuman boundary and to rigorously decentralize the human being. It is true that silhouettes can be recognized as such only through the human gaze and acquire meaning through human perception. Nevertheless, the posthuman conception of silhouettes makes it easier to consistently consider the agential cuts made in an analysis and prevent researchers from falling back into human-centered analysis. Making (the properties of) human and nonhuman indeterminate is possible through the layout of the silhouettes: color schemes, textures of materiality, designs, and shapes do not allow clear conclusions to be drawn about the entities involved in a formation. Latour’s (2000) idea of symmetry is taken literally here. It prevents attributing a different status to human and nonhuman. By considering these methodological ideas, we can demonstrate that the dualism of human/nonhuman is not always empirically valid.

The silhouettes analysis does not use written data such as interview transcripts, but rather image details, which are made into silhouettes and then examined. Taking Barad’s idea of material-discursive practices seriously, silhouettes analysis is a good complement to the posthuman analysis of verbal methods. For while these analyze articulated contents—taking into account gender, sociocultural background, and also facial expressions, gestures, and sometimes emotional states—and examine them as expressions of assemblages of human and nonhuman (e.g. Höppner, 2015a, 2015b, 2017), in silhouettes analysis the formations and the movements or movement models in which a phenomenon takes place are of interest and are thus visualized and analyzed. Hence, interview material makes it possible to hear and to read how a phenomenon is constituted; silhouettes make this process visible. This makes it all the more important to train one's own researcher's gaze, but also to irritate it in order to stimulate new ways of seeing. Silhouettes analysis enables a much more thoughtful way of seeing: through the use of silhouettes researchers can gain new insights and see something different from what we typically see when we think we know what a thing looks like.