Hidden racism and structures of power in the images of Ebola 2014

Epidemics are considered paradigmatic states of emergency and humanitarian scenarios. Thus, humanitarian conceptualizations are negotia-ble through the very practices appearing in this type of emergency. This paper aims to investigate this process in relation to the 2014 Ebola outbreak, an event that is considered a global threat and an intolerable humanitarian situation. We analyze how the definition of what can be understood (or not) as humanity was constructed through visual representations produced by social media. We will also discuss how the definition of humanity was negotiated through dimensions such as the spectrum of visibility, the distribution of agencies, the affections intended to provoke, and the imaginaries defined. This paper is based on an empirical semiotic analysis of hundreds of images from the 2014 Ebola epidemic and 15 focus groups and individual interviews, performed over a span of one year, discussing images from social media. UAB psychosocial intervention and in biopolitics, decolonial and feminist theories. Her master’s thesis analyzes how humanity is negotiated in epidemics and health disasters through the images produced in these scenarios, using the specific case of Ebola 2014 outbreak. Currently, she is interested in critical psychology and new approaches to treat mental health crises, especially Open Dialogue. actor-network sociotechnical a) The remake of Serres’s notion “extitution” Atlas. It in a our near future, but without content or definition (Serres way of writing), and it, suitable this will grow strong and in all directions”; the notion of cinepolitics (kinepolitics) that used in several papers to describe a new governmentality practice focussed on the control and administration of our daily movement. cinepolitics is interesting expand the limits of biopolitics. his intercessors (those authors


Introduction
Epidemics and pandemics are not just simple medical or biological issues. When a pandemic is declared by the World Health Organization (WHO), local governments, states and global institutions mobilize experts and resources to handle the situation. This mobilization includes medical assistance, vaccines, hospitals and epidemiological measures. However, these biological catastrophes come with social and political laws, interdicts, new practices, global fears or unexpected habits. COVID-19 is a clear example of why not only biological sciences but also social sciences are needed to deal with bio-emergencies and the associated potential consequences. Medical actions have a remarkable impact on social measures, such as the lockdown that has been imposed in several countries. Some of the main consequences observed are high rates of unemployment, economic crises and severe psychological damage. However, unlike COVID-19, the 2014 Ebola outbreak exhibited a different magnitude, scale and symptoms, and different global strategies were adopted to combat its spread. The common epidemic nature of COVID-19 and the Ebola outbreak has allowed us to retrospectively analyze the Ebola outbreak and extract lessons to be applied to the current and to future pandemics.
As pointed out by several authors, epidemics are historical phenomena with relevant consequences on the social imaginary of cultures (Watts, 1999;Zylberman, 2013). Epidemics can even be considered as a political momentum through which entire nations or cultures transform their structure (Caduff, 2015;Mason, 2016). Several studies have also considered them as a potential field to analyze micro-social processes, such as the constitution of power relationships on life, the definition of new human collectives or the relationship between experts and laypeople (Bergman, Kawachi and Glymour, 2014;Chermans & Coons, 2015;Lakoff, 2015). Another specificity that makes epidemics relevant to social sciences is that they are special moments that reveal and maintain what Bourdieu (1979) called symbolic violence, a type of silent and invisible violence, making it difficult to detect, even for people experiencing it. In this vein, Michael Foucault analyzed the scientific discourse as a key maintainer of social inequality, rather than it being the neutral knowledge that can discover and describe this violence or inequality (Foucault, 1985(Foucault, , 1994a(Foucault, , 1995b. Following this path, Butler (2004) and Haraway (2013) showed how science and technology reproduce gender differences or insuperable distances between experts and laypeople through their practices. Similarly, Rose (2007) considered the ways in which the development of biomedicine has affected the constitution of new power relationships, racial politics or new mechanisms of crime control. Moreover, Harding (2011) re-evaluated the conventional accounts of the west's scientific and technological projects, re-thinking the strengths and limitations of the knowledge traditions of non-western societies and assessing the legacies of colonialism and imperialism. Finally, Mbembe (2008) and Fanon (2008) claimed that racist policies are implemented and naturalized in the name of the 'common good'. It is worth noting that epidemics are interesting moments for social sciences because they deploy states of emergency, such as special situations in which society is re-defined in terms of its normative, moral and economic structure. A state of emergency is a moment in which the state imposes all of its restrictive power on its citizens. In this sense, Agamben (2004) claimed that the de-humanization of human collectives is a potential consequence of a state of emergency, with the Holocaust being the most significative event explaining this phenomenon. Notably, Agamben referred to this perverse mechanism in the core of democracy: the conversion of some human beings (bios) into de-humanized humans (zoe). This dehumanization of some collectives or individuals generates prejudices, indifference or even hate towards them, and, in the long term, allowing systematic violations of human rights. As an example of this phenomenon, Seebach (2016) analyzed how an epidemic outbreak (scabies) in a migrant camp in Germany generated fear, hate and anger towards this collective. This is considered one of the main reasons why the Ebola epidemic is an ideal case study to analyze the creation of other human collectives, the impact of different affections (e.g. fear and threat) and the weight of colonialism and racism. All of these elements af-fect how the humanity of suffering individuals, as well as our own humanity, is in continuous negotiation within the force field of living. This paper is structured as follows. A detailed analysis on the concept of 'humanity' is first presented, followed by a contextualization of the 2014 Ebola outbreak. Then, insights into research using images are provided, as well as their epistemological and philosophical implications. This is later followed by a description of the data collection and research analysis. The following categories are proposed to explore the negotiation process of the meaning of humanity in biological emergencies: a. what can or cannot be seen, b. the distribution of agencies, c. the affections they intend to provoke and d. the imaginaries or myths depicted/activated. Our paper finally concludes by showing how Ebola-related images convey the dialectic threat-humanitarian affection as a central message. Thus, it is not exaggerated to claim that images of biological catastrophes, such as the Ebola outbreak, speak volumes about a new and subtle type of racism.

Humanity under a humanitarian government
Epidemics are a heterogeneous field in which national and international politics intertwine with ethics, laws and humanitarian work (Lakoff, 2010). Most theories tackling the topic of humanitarianism emphasize that humanitarianism is not a mere exchange between two or more social actors, but rather a whole form of governance with special relevance in crises or emergencies. One of the factors making humanitarianism a double-edged sword is that it is performed as an apolitical and ahistorical activity (Ticktin, 2014). By placing human suffering at the centre of its mechanism, humanitarianism blurs or at least displaces deeper understandings of inequality, such as political and historical analyses. The intervention of different agents (wide webs of experts, health workers, health assistants and non-governmental organizations), as well as the nexus between them, causes humanitarian work to be apolitical, with the only common objective being to avoid human suffering in general (Lakoff, 2010). Therefore, socio-political problems become psychologized and considered in terms of personal distress.
However, most researchers agree on the fact that humanitarianism is only apolitical on the surface; for example, when it is presented on mainstream media or when it is spoken of in a generic discourse on human rights advocacy. A deeper understanding on the topic reveals that it is in fact linked to global mechanisms of neo-colonialism, surveillance and power dynamics. So far, sociological scopes labelled humanitarianism as 'the left hand of the empire' (Agier, 2010), claiming that the same agencies that create and maintain global inequalities are the ones undertaking humanitarian work. According to Lakoff (2010), humanitarianism is the key for international bio-security, as it does not guarantee global health but rather guarantees global surveillance. Thus, it is possible to define two main western ways of establishing global health bio-surveillance mechanisms: 1. global bio-security aiming to protect the population from epidemics originating in undeveloped countries, even before they emerge, and 2. humanitarian bio-medicine that addresses diseases generally affecting undeveloped countries.
Thus, it is possible to hypothesize that surveillance and bio-security take place at the centre of international operations. Both regimens were observed to converge during the Ebola outbreak because, aside from being a target of humanitarian intervention, the spread of the virus made the outbreak an international threat. This international strategy is particularly relevant in the case of epidemics, in which surveillance is perceived as a necessary element to prevent diseases from spreading. In addition, global surveillance is connected in genealogical terms with the government's strategies of colonialism (Guardiola-Rivera, 2008), especially when tackling humanitarian catastrophes occurring in the so-called 'third world'. As will be further analyzed through images, the fact that the 2014 Ebola outbreak originated in Africa is entangled with colonial history and plays an important role in the dehumanization process.
In this paper, the main implication of humanitarianism that we will analyze is that it operates directly on the idea of 'humanity'. As suggested by Fassin (2010), humanity refers on the one hand to the shared condition of human beings (mankind) and on the other hand to the affection that people feel towards their fellowmen (humaneness). Humanitarianism produces certain types of subjectivity to being governed, due to vulnerability, social status or individual or collective suffering, generating a kind of subjects susceptible to being governed and other subjects susceptible to being humanitarian. The importance of this process is that it is a moment for re-conceptualizing the humanity of humanitarians and their targets alike. In conclusion, these factors make humanitarianism an ideal scenario for analyzing the relationship between different agencies and the construction of other human collectives.

Depicting humanity within a catastrophe: images of pain
In western cultures, images have been used to communicate the sacred word to a widely illiterate population, where depicted scenarios played an important role in the propaedeutics of living (how people should act, what people should fear, what is forbidden, etc). Understanding the power of such images is particularly interesting in the case of images of pain, such as martyrs being tortured or sinners receiving punishment. Analyzing the mediation of different devices over our affections, Blanco (2016) stated that images, music and performances have always been orchestrated to suggest specific affections to the audience. For example, in the case of sacred images, the people surrounding the suffering individual are key to interpreting the picture. Sorrowful characters in a picture depicting a saint's martyrdom indicate that the viewers should feel pity and that pain is unfair. On the other hand, judgmental or pleased faces surrounding a sinner communicate a feeling that the punishment inflicted is deserved. This is one of the ways in which images have powerful implications over the 'humanity' of the depicted subject, as well as over the viewer's affections. In other words, images propose grammars on how to live or die in a way that makes sense within a shared socio-cultural context (Blanco and Castro, 2005). Nevertheless, images should not be understood as an isolated mechanism. Debord (1983) stated that a spectacle is not a set of images, but rather a social relationship mediated by images. According to Rancière (2010), images are embedded in a strategic mechanism that determines what we should see and how we should see it. Particularly in a media context, one of the consequences of images is that they create the illusion of a simple and homogeneous perspective of complex social realities. Additionally, the media often reproduce the mainstream bias, contributing to creating and maintaining colonial imaginaries, prejudices and discrimination, among others. In this vein, Fassin (2010) stated that social protests found a new visual display of humanitarian catastrophes in the mid-20th century. Images of blood and horror were replaced by faces and gestures of suffering. This shift originally aimed to provoke affections and moral responses in the spectator and ultimately, denounce the intolerable. However, the effects of these images on the humanity of the individuals were controversial. For instance, Calain (2013) questioned whether the pictures of suffering people respect their privacy and dignity. Through his analysis of pictures in humanitarian medicine, he concluded that medical ethics are often not respected or are respected only in catastrophes that occur in developed countries. In the same vein, Barbie Zelizer (2005) analyzed how the visual depiction on the war in Afghanistan shaped the public opinion regarding this specific event, highlighting that there are always a several steps in decision-making on what is worthy of depiction. She argues that the strategic visual coverage of the event was used to make sense of a war in some way: in the case of the Afghanistan war, the images of death were used to maintain public support.
As Sontag outlined in her brilliant analysis 'Regarding the Pain of Others', "photographers (...) have become increasingly concerned with the issues of exploitation of sentiment (pity, compassion, indignation) in war photography and of rote ways of provoking feeling" (Sontag, 2003:71).
This statement implies that by attempting to 'humanize' spectators by showing them explicit suffering may result in the objectification and de-humanization of the victims. Diéguez (2016) performed an analysis of the iconography of pain, pondering over whether horror can or cannot be represented and if so, what the 'correct distance' and 'correct use' is for these kinds of images. These are also the questions that we will address in this paper. Hence, we would like to understand what characteristics make certain images of pain susceptible to being understood such as a common place, a community to empathize with one another or instead, just a gateway to normalize pain and dehumanize suffering individuals. Therefore, in our opinion, how images depict humanity deserves a thorough analysis. Considering the above-mentioned research on humanitarian images, the following analysis categories are the ones proposed for exploring humanity: a. what can or cannot be seen, b. the distribution of agencies, c. the affections they intend to provoke and d. the imaginaries or myths depicted/activated. This analytic procedure will be applied to the case of the 2014 Ebola outbreak.

Contextualizing Ebola
To understand the subject we are dealing with, a brief contextualization of the 2014 Ebola epidemic is necessary. The 2014 Ebola outbreak is considered the largest Ebola epidemic in history, with 28,652 cases and 11,325 casualties. This outbreak was officially declared on March 23rd, with several cases reported in Guinea, and soon spread widely to Liberia and Sierra Leone, with a smaller number of cases in other African and European countries.
As mentioned by Goffman (1974), every social phenomenon is framed by a socio-cultural apparatus comprising cultural viabilities, discourses, practices, media normatives and so on, and an attempt to isolate the former from the latter is always a sterile effort. Thus, mass media tended to compartmentalize the Ebola outbreak by focusing on medical or geographical aspects. However, this was a complex phenomenon that involved not only the virus itself, but also its contagious nature. As put forward by Wilson, we must understand 'Ebola as a discourse, not just as a disease ' (2015:5). In line with this, in his research on the Ebola epidemic, anthropologist Niang (WHO, 2015) stated that 'it is not just a biomedical problem. It is a political and social fact, developed within a historically postcolonial field. It is the child of a history of structural and multiform violence'.
Unlike other humanitarian disasters, such as famines and natural disasters, epidemics are considered a potential threat because of their ability to spread, which is not the case (at least not directly) of famines. Notably, COVID-19, H1N1, AIDS and the successive epidemics of cholera south of the Sahara all have similar characteristics in terms of this potential dangerousness factor. The geopolitical location of the source of epidemic is a key element to analyze the prejudices associated with this continent, which are activated because of the fear/alarm at the epidemic. This specific racism that emerges in media and public discourses must be understood within the historical heritage of colonialism, as it has been largely pointed out by black intellectuals such as Mbembe (2008), Fanon (2008) or Angela Davis (2011).
This complex phenomenon, 'epidemic' or 'pandemic' plays an important role in the visual display produced by social actors, such as mass media, social movements or political institutions. In the case of the 2014 Ebola outbreak, this fact was considerably more noticeable than in other similar outbreaks. For instance, during the H1N1 outbreak in Europe in the 2000s, the mass media only showed graphics depicting maps of the spreading of the virus. However, in the case of Ebola, images of dead people on the streets were used. Therefore, images are another relevant dimension to analyze in order to understand the development of an epidemic and its social and political consequences. In this sense, our research group focused on this topic and found an enormous amount of visual material showing specific differences from prototypical images in other epidemics, such as H1N1. Notably, the 2014 Ebola outbreak is considered a paradigmatic case in which the visual representation factor (photographs, prevention campaigns, explicative videos, scenario/stage re-creation, etc.) had a significant effect. This type of representation in the media caused several repercussions (images of infected people, bodies piled up and medical protective suits, implying the dangerousness of the disease, followed by a detailed description of the symptoms accompanying this infection): 'Though clinically inaccurate, the language of liquefying or melting organs is viscerally terrifying' (Wilson, 2015:17). Nevertheless, using images in investigations is not a direct or simple job. Generally, gathering shocking pictures and commenting on what is shown in them is not enough to investigate the effects of these images. Instead, dealing with such images implies working with a whole epistemological and philosophical background or tradition that must be known and made clear.

Researching with and about images
Some authors maintain that "the visual is central in the cultural construction of contemporary social life" (Jenks, 1995;Rose, 2012), also referred to as the visual culture in human sciences. From the works of GuyDebord (1983) on the society of spectacle to the contribution of Virilio (1994) on the vision machine in which all of us are contained, or from the interest for history of art due to the evolution in the production of images to the philosophy that maintains that ocularcentrism is the dominant trait in western societies, all these contributions show the central role that images play in our lives. Such concern has also been demonstrated in social sciences in what is called Visual Research Methods. Generally, three elements characterize the aforementioned concern (Barnard, 2001; Rose, 2012). First, it is considered that, through images, we can speak of social differences that make up our social reality. Analyzing images introduces us to the differentiating scales and traits of our social structure. Second, it is postulated that it is very important to examine how images show these social structures and how they must be viewed. Finally, it is assumed that there are systems or ways of looking. Our eyes do nothing but look at particular assemblies of wisdom, institutions and historic trajectories.
More recently, within the Visual Research Methods, a movement called multimodality or multimodal discourse analysis (MDA) has achieved some relevance. Notably, MDA is an emerging paradigm in the study of the discourse that extends from the study of language per se to the study of language in combination with other resources, such as images, scientific symbols, gestures, action, music and sound (O'Halloran, 2011:120). Here we highlight one postulate within this movement: the idea of working simultaneously with texts and images and considering them a semiotic reality in which neither is reduced to the other and both maintain their essence or ontological statute. Several authors, such as Rose (2012), called this capacity inter-textuality and discussed how different discourses are inter-related, either verbally or visually or in a mixture of both. Other authors, however, prefer to discuss inter-semiotic or inter-modal de-relationships (Jewitt, 2009).
Notably, Rose (2012) offered a slightly different proposal to the previous one that is considered interesting to review. To start with, she distinguished between a process that she calls Analysis of the Discourse I and another called Analysis of the Discourse II. Both processes can be complementary but are not necessarily so, as they have sufficient autonomy to construct valid tools of wisdom production by themselves. In the Analysis of the Discourse I, we find the following phases. First, the general information of the images and text about the topic that we are interested in must be consulted. The broader it is, i.e. the better and the more sources covered, the more interesting the analysis will be. Second, the images must be analyzed using iconography. Following the classic work of Panofsky (1957), Rose proposed dividing this analysis into three phases: primary or pre-iconographic, secondary or iconographic and intrinsic or iconological. The difference between these phases resides in the depth (from smaller to greater) with which we gain access to the significance of an image: First from a simply visual and descriptive recognition in the pre-iconographic phase to a social and cultural interpretation in the iconology phase. Second, it is essential to respond to the questions of who, where, when and for whom the images have been expressed to define the social context or framework in which the discourse is delivered. Lastly, it is necessary to analyze what kinds of aspects in images go unnoticed. Analysis of the Discourse II helps to understand images as a direct product of certain institutions and conditions of knowledge and power. Thus, in this sort of analysis of discourse, images are linked by their significance with power and institutional structures. Moreover, this analysis shows that the production and significance of such images are scientifically relevant.
It has been reported that the MDA enriches the classical analysis of nonlinguistic discourses, such as images and sounds. In this sense, it has been established that the interaction between different expression modalities is a key factor in the understanding of our social reality. Notably, MDA puts forward three analytical propositions. First of all, analyzing images allows us to examine social differences. Secondly, images help to establish how we contemplate certain aspects of our social reality. Finally, these ways of seeing are intrinsically linked with the way of living and experiencing reality. Therefore, the MDA is considered an adequate tool for analyzing phenomena such as epidemics, in which images, texts, figures and graphics seem to be a seamless web.

Data collection
This research was part of a broader two-year European project (2014-2016) addressing lay discourses in biological catastrophes. Two different types of data were used. First, images from the 2014 Ebola outbreak were collected from journalism, WHO bulletin, scientific protocols, guidelines and informative warning panels. A semiotic analysis was conducted by the research group, following the previously explained theoretical underpinnings of multimodal discourse analysis. In the first place a "visual and descriptive recognition of the iconography" was performed, followed by a "social context of framework". The output of these analyses threw significant insights that made it possible to "study what kind of aspects in the images are common and go unnoticed", in order to understand the power dynamics in the images that will be described in the results section.
A small sample of pictures have been selected to be shown in this study to conduct an analysis. Second, a sample of the most representative images were presented in focus groups (FGs) and individual interviews. In total, 10 FGs and five individual interviews were conducted in Barcelona, Spain, with a total of 80 participants. The use of this qualitative methodology was suitable for the objective of the study: we wanted to capture the general discourse on this topic, for which we recruited a random sample of participants of different ages, socioeconomical backgrounds and with or without scientific knowledge. At least two experienced researchers conducted each group interview. The following ethical procedures were implemented in the research process: a. in the semiotic image analysis, permission was obtained for the use of non-royalty-free images; b. in the FGs and interviews, an informed consent form to participate, explaining the research objectives, was signed by all the participants, and withdrawal from the discussions was allowed at any time point.

Analysis: negotiating humanity
Generally, we have analyzed our images while paying attention to four dimensions: a. what can or cannot be shown, b. the distribution of agencies, c. the affections they provoke and d. the creation of imaginaries.
These four categories were obtained from studies in the literature on the semiotic analysis of images, describing: a. what can or cannot be shown in a specific event, b. how images result in specific distributions of agencies, c. what sort of affections a specific image can provoke and d. how a set of images can create or evoke a social imaginary.

Spectrum of visibility:
What can or cannot be shown Significant differences were observed in the spectrum of what is visible in the Ebola outbreak compared to other epidemics. One of the relevant features that was observed in the pictures is the large number of victims and sick people, displayed in a way (faceless, semi-naked, lying on a dirty ground) resembling a dead animal (see Figure 1). As outlined by Sontag (2003), the western visual depiction of pain tends to hide sacred elements and show non-sacred ones. Hence, it can be understood that images showing sick or deceased people as a result of Ebola are considered less sacred because they can be shown. This finding has powerful implications for 'humanity'. As highlighted in this paper, images are cultural devices used by people to articulate and understand their own experiences. This is why when these types of images are shown, the first idea that comes to the FGs is that the way in which these people live and die is different from the western way. One of the features that reinforces this idea is the apparent indifference of the people surrounding the victim.

Source: Abbas Dulleh
Notably, all the participants focus on the differences between the western world and the 'under-developed' world, creating a gap between the humanity of the people in the pictures and their own humanity.
"Well, there is no healthcare, no prevention… they are used to seeing dead people all the time".
This feature is common in non-western humanitarian disasters and contributes to the creation of an inferiority imaginary of the represented cultures (Fassin, 2012). Although this study is not a comparative one, we do propose a quick comparison with prototypical visual representations of H1N1 (an outbreak mainly affecting first-world countries) to understand the structural differences between Ebola and other epidemics.
Most images show the geopolitical distribution and protocols used to deal with the disease (Figure 2). For example, the map above (Figure 3), provided by the WHO, shows that the spread of the epidemic was highest in the USA, Australia and parts of Europe, as well as in some countries in South America, such as Chile and Argentina. However, no representations of infected people are shown. Only a small number of images with people are shown, mostly depicting experts or individuals wearing masks, but not showing whether they are infected or just regular people following preventive protocols. None of the images shows dying bodies on the ground or violent manipulation of patients. As we explained, we are using a multimodal discourse analysis, which means that all visual inputs are going to be treated as images. Therefore, maps and infographics are analyzed in the same way as photographs because they also contain relevant information about power structures. These differences in the images shown are not innocent. Indeed, they are related to the construction of humanity and, as will be analyzed in the following point, to differences in the subjects and agencies depicted.

Distributing agencies: We are all subjects, but some of us are more subjected than the others
The semiotic analysis of the images shows that two types of subjects appear in most of the images: 1. health workers wearing protective clothing and 2. people belonging to the wide category of infected/dying individuals.
By analyzing the actions represented in the images, it can be observed that these actions are performed by health workers, with the victims as passive parties. As depicted above, contact between the health worker and the infected/possibly infected person is often mediated by objects (Figure 4). Looking at this image, it can be seen that it depicts symbolic violence based on the supremacy of the disguised health worker, with the woman having an ambiguous, completely passive posture.  Notably, the WHO summarzed this view of health workers in one of their bulletins after the outbreak: "By overcoming their fears and having the courage to work with Ebola patients, health workers have emerged as the heroes in this epidemic" (WHO, 2015).
Our results here are in line with that argument, but in fact we consider the situation to be way more complex. Humanitarian workers are a subject of contradiction: they can either be irresponsible and pose a problem for the whole country, or they can be seen as 'brave heroes' and even be glorified. This is like the historical figure of saints, whose superior humanity status allows them to look down on others to save them. However, unlike saints, humanitarian workers save others not by touching and counselling them, but rather by keeping a safe distance from them. Their objective is to save people without necessarily acknowledging them. Therefore, these 'new saints' save people not by proximity but rather through their ability to keep a safe distance from the subjects they are trying to save. Being unable to identify the person in the costume makes it challenging to empathize with him/her. The second aspect seen in the images is the sick people, who appear to be mostly black, half-naked bodies whose faces are commonly not shown (see Figure 6).
Some images even show that these humans are being handled (sometimes violently, as shown in Figure 6) by individuals in protective clothing, but neither of them is identifiable. The individuals represent the majority of sanitary workers. This lack of privacy in the images (this depiction of the deceased or sick people in houses or hospital rooms, which should be private) leads us to an imaginary of danger, threat and a shared illusion that they are being treated this way for a reason, for the common good. The 2014 Ebola outbreak is a humanitarian catastrophe that had a double effect on western countries. First, it provoked feelings of intolerance as a result of the extreme suffering and violation of human rights and global health assistance. Second, it provoked feelings of threat stemming from the fear that the catastrophe would spread and threaten our social wellbeing. Our results show that these feelings of intolerance and threat are not confronted per se but rather shifted towards other ways of affection. Although the first feelings that the FGs associated with the images were fear, threat and danger, these feelings soon shifted to other less negative issues. Notably, a similar mechanism was observed when the topic of intolerance was brought up in the group discussion. Feelings are managed in different ways, which is an essential aspect for understanding the mechanism of dealing with these kinds of images. The following are some of the most common ways to deal with these feelings.

Embodying the threat in the infected individual
Usually, feelings of threat shift to feelings of blame and the sole responsibility of the individual, such as in the case of humanitarian workers, or instead to the issues of immigration or uncivilized living conditions (when talking about African citizens). Migrants are often viewed as an uncontrollable flux of viruses that pose a threat. This displacement of feelings is essential when dealing with the pain of others. "The infected ones… those are people coming from outside. A cure must be developed in their countries of origin." "Why do they come? They are endangering a whole population [the Spanish one]." At this point, there is a debate regarding what is politically correct to say. The meaning appears to be the same (black individuals migrating to Spain, whether infected or not, still represent a threat), but the displacement of accusation transforms this subject into a non-agency figure. In this sense, rejection does not directly focus on these immigrants, but rather on their conditions and ways of living. In other words, although this individual is not infected, he/she is exposed to the conditions that might get him/her infected. The centre of the image shows a covered stretcher that may contain a corpse. Notably, without any further information, it is kind of difficult to understand this scene. However, a detailed look revealed that these people were praying, an observation that neither of the FGs made. This may be because praying in the western concept is not often performed under these circumstances (i.e., outdoors, in the forest, over an unrecognizable corpse lying on the ground). Some of the focus groups used this image to claim that the 'suffering' individual' should be cared for (meaning that its corpse should be treated with dignity and not left abandoned in the dirt), but when asked about the general care of immigrants the general view turned into 'uncontrollable fluxes of immigrants' that should be deported from our countries. "There must be a rigorous control of people coming from 'these' countries because they are exposed to many diseases. As long as there is no control we are exposed to getting contaminated by any plague."

Displacing the threat towards control and surveillance
Notably, it seemed that feelings of pity prevented the participants from blaming the suffering victim, making blaming anonymous 'fluxes of immigrants' an easier choice.
There were some eradicated diseases which are spreading again because of uncontrolled migrations…"

Displacing the intolerable towards humanitarian affections
Instead of discussing the issue of what is intolerable in the Ebola case (Why was it intolerable? What could have been done to prevent it from happening? What can we do to avoid similar events from happening again?). The discussion was oriented towards vague cliché critiques of global inequality. "They [regarding migrants] come without any vaccine control, sanitary control… people, south Africans, come by boat and are released in the cities… and these people do not have the controls we have here, which is unlucky for them, and they bring diseases…" "Those poor people… that shouldn't be happening." "I haven't seen a white man like this, left to die on the dirty soil." Discourses that highlight differences between 'us' and 'them' also create a gap between our and their ways of living, dying and suffering. All the depicted mechanism does is increase the differences between 'we' and 'them' because it asserts that we have different human conditions. By removing the agency of the victim, he/she can be perceived as just a recipient that embodies the threat or a nameless victim, not as an equal.

Imaginaries and myths (Africa, the viral jungle)
Notably, the first thought regarding the source of a threat in an epidemic, is usually developing countries, with most FGs pointing at African countries. In some cases, South America was also mentioned, mainly related to the Zika virus. Though not explicitly in all cases, Africa is considered a viral jungle with a rudimentary way of life. The chasm between the 'civilized us' and the 'wild unknown others' stems from the idea that their ways of life are all too different from ours. This African imaginary is reinforced by multiple mediatic representations, cultural myths and social discourses. Images related to the Ebola outbreak (see Figure 9) depict a context of complete chaos and primitive practices: shoeless people, indifferent-looking people passing by corpses, people without proper protective clothing close to infected people (despite the high rate of contagion), ancient rituals and shamanism. The points presented by the participants of the FGs comprised the old prejudice of the 'dangerous African jungle'.

Exception image
The image above shows a doctor talking in the upper-left corner (health expert), a graphical representation of the Ebola virus and health workers, all of them surrounding a half-naked individual whose face is not shown. No face-to-face contact exists between the victim and the health workers. This picture is highly ambiguous because it contains strong elements displacing our attention: the doctor, who represents the medical authority and expertise, and the virus, representing the danger. As shown in previous images, the victim shows the same passive characteristics. The visual display of the Ebola outbreak recreates a continuous state of exception in which history and politics are suspended to maintain common safety. These images depict a state of emergency, as they provoke fear and panic, suggest a threatening future and prompt us to protect ourselves. These types of pictures are called exception images, inspired by the work of Ileana Diéguez, who created concepts such as the Medusa image, tomb image and trace image for her research on images of catastrophes. Therefore, we suggest that images like the one depicted above reflect the core concept of the state of exception, especially the suppression (or potential suppression) of humanity, and help create new types of racism.

Discussion
Recently, many studies have shown that the social and political rejection of explicit racism in developed countries resulted in new and subtle types of racism (Zizek, 2008, Bauman, 2012. Studies have also shown that the micro-mechanisms operating in such cases can offer a deeper understanding on how prejudices are being maintained, being thus a useful tool for re-conceptualizing racism. The semiotic analysis addressed different micro-mechanisms regarding the sophisticated way in which Ebola-related images are presented: the spectrum of visibility legitimates the infra-human portrayal of others (beast-like, faceless, and uncivilized) in a scenario of fear and threat. These characteristics provoke a threat and humanitarianism by reinforcing colonial imaginaries. In addition, most of the images evoke cultural imaginaries of fear, threat, danger, etc; which may be displacing the discussion on human rights. As shown, most images depict a state of emergency: a safe distance from others, painting a picture of a catastrophic future that demands self-protection. On the other hand, humanitarian affections also incorporate a potential type of racism when people are victimized because of their different human conditions and are in dire need of our civilized help.
In this sense, racism emerges in the imaginaries raised by images: not only our laws, but also our affections, empathy, compassion and conception of justice become re-conceptualized because of the emergency caused by Ebola-related images. These violent images shown in the media are a contemporary memento mori, in which bodies are used to communicate fear. Therefore, it can be understood from our analysis that Ebola is not being fully depicted. Instead, it is just a metonymy that represents de-contextualized fear/ terror. In other words, photographs showing naked corpses lying in the middle of the road do not depict the whole picture, the life of that human being and the pain and suffering of his/her relatives.
However, these results do not mean that events like the Ebola outbreak should not be depicted. As pointed out by several contemporary authors, we become an accomplice to horror if we assume that it cannot be represented. In her analysis on the iconography of pain, Diéguez claimed that censuring images of terror can lead to the concealment of socio-political violence and block one of the possible ways by which such injustices can be denounced, a point with which we strongly agree. Our research aim was to analyze these types of images carefully and critically, to detect the elements that are contributing to de-humanization and colonialist discourses.
In contrast, some forms of photojournalism advocate for a different representation of humanitarian catastrophes, such as Gervasio Sanchez's collection. This photographer portrays victims of violence by attaching their names and biographies, aiming to reveal the real person behind the label of 'victim'. This form of depiction results in images that are different from the typical images in the media, such as what happens after a catastrophe, how people live and how they survive (e.g. a very affectionate scene of a mutilated woman breastfeeding her baby or a family holding a photograph of a missing member). In that sense, I survived Ebola (http://isurvivedebola.org/) is another example of a project that aims to maintain the humanity of the victims: it aims to present a different picture by depicting Ebola survivors as well as their history and everyday life after the outbreak.

Conclusion
Throughout human history, there has always been a gap between viewing other people as a threat and viewing them as a humanitarian target. Moreover, there is also a gap between the psychologization of a catastrophe (by looking into the very threat or suffering in an individual) and the wider scope of blurring the individual in pursuit of structural violence and power relations. In both cases, humanity plays a central role. As we have exposed, fear and humanitarian affection are the main reactions to Ebola-related images. In this sense, it is not exaggerated to claim that images of biological catastrophes, such as the Ebola outbreak, speak volumes about a new and subtle type of racism. The aim of this study was not to find out whether the visual depiction of the Ebola outbreak corresponds to some events of the catastrophe (i.e. Did all Ebola cases take place in scenarios like those depicted? Did all humanitarian and health workers treat the infected people with the same symbolical violence?). Rather, what we wanted to show through our analysis is how the micro-mechanism of representation creates some operations over humanity, as well as the consequences they have on discourses and social imaginaries. In the same way that Foucault (1985) and Haraway (2013), as well as many other researchers, denounced the effect of science on social inequality, we should start wondering how neutral media coverage fuels and legitimatizes racial inequality in a subtle way.
Finally, we would like to include our research as a specific case within a wider scope: how western democracies can cause and cope with humanitarian catastrophes and systematic violation of human rights. As mentioned above, Agamben (1998Agamben ( , 2004 wrote about the conversion of human beings (bios) into de-humanized humans (zoe). Our research here highlights one specific way of de-humanization that does not clearly contradict our civilized way of life.
"It is rather to try and understand once and for all why democracy, at the very moment in which it seemed to have finally triumphed over its adversaries and reached its greatest height, proved itself incapable of saving zoē, to whose happiness it had dedicated all its efforts, from unprecedented ruin" (Agamben, 1998).