Joining the conspiracy? Negotiating ethics and emotions in researching 
(around) AIDS in southern Africa  
 
Nicola Ansell and Lorraine van Blerk1
Department of Geography and Earth Sciences 
Brunel University, Uxbridge, UB8 3PH 
nicola.ansell@brunel.ac.uk; lorraine.van.blerk@brunel.ac.uk  
 
Abstract 
AIDS is an emotive subject, particularly in southern Africa. Among those who have 
been directly affected by the disease, or who perceive themselves to be personally at 
risk, talking about AIDS inevitably arouses strong emotions - amongst them fear, 
distress, loss and anger. Conventionally, human geography research has avoided 
engagement with such emotions. Although the ideal of the detached observer has 
been roundly critiqued, the emphasis in methodological literature on 'doing no harm' 
has led even qualitative researchers to avoid difficult emotional encounters. 
Nonetheless, research is inevitably shaped by emotions, not least those of the 
researchers themselves. In this paper, we examine the role of emotions in the 
research process through our experiences of researching the lives of 'Young AIDS 
migrants' in Malawi and Lesotho. We explore how the context of the research gave 
rise to the production of particular emotions, and how, in response, we shaped the 
research, presenting a research agenda focused more on migration than AIDS. This 
example reveals a tension between universalised ethics expressed through ethical 
research guidelines that demand informed consent, and ethics of care, sensitive to 
emotional context. It also demonstrates how dualistic distinctions between reason 
and emotion, justice and care, global and local are unhelpful in interpreting the ethics 
of research practice. 
 
Key words: emotion; ethics; ethical codes; HIV/AIDS; Africa 
 
Introduction 
The past two decades have witnessed a plethora of articles discussing the ethics of 
conducting human geographical research. The roots of this concern can be traced to 
humanistic geography and the rejection of the ideal of objectivity by many 
geographers. Recently, however, Widdowfield (2000) has pointed to the fact that 
                                                 
1 Nee Young. 
 1
despite recognition of subjectivity and efforts to engage reflexively with research, 
almost all accounts of the research process omit discussion of the influence of 
emotions on that process.  
 
The fact that emotions are excluded from reports of research does not mean that 
they are not taken account of in the design and management of research. We would 
suggest, however, that the primary emphasis has been on minimising the 
‘interference’ of emotions in the research process and in particular the avoidance of 
situations of heightened emotion within research encounters. This avoidance may in 
part be attributable to the persistence of a tendency to privilege ‘masculine’ ways of 
knowing: 'the Academy appearing to remain wary of acknowledgement and 
expressions of emotion' (Widdowfield 2000, 200). Avoiding emotional encounters is 
not, however, simply an endeavour to remain objectively detached. It is also related 
to nervousness about intervening in people’s lives in a negative way: to lay bare a 
person’s emotions is seen as an intrusion into their privacy. Arousing negative 
emotions in research participants seems counter to injunctions to researchers to ‘do 
no harm.’ 
 
Yet the relationship between emotions and ethics in the research process is much 
more complex than this. The first part of this paper examines contemporary literature 
on the relationship between ethics and emotions. On the basis of the co-dependency 
between emotion and cognition, we argue that emotion and reason are not binary 
opposites, and that this has important implications for ethics. We explore the tensions 
between an ‘ethics of care’ grounded in interpersonal emotional relations and an 
‘ethics of justice’ based on universal principles of fairness, deconstructing the 
mapping of these ethical modalities onto a local-global binary. We argue that taking 
account of emotions through application of an ethics of care is appropriate to 
research encounters with people in Third World settings, but may conflict with ethical 
codes which seek to impose universal ethical principles. The second part of the 
paper applies these ideas to specific research conducted among AIDS-affected 
people in southern Africa, which inevitably required us to address difficult issues 
concerning both emotions and ethics. The (global and local) discourses that surround 
AIDS in southern Africa may provoke a range of uncomfortable emotions among 
researchers and research subjects. We outline how we managed the research in 
order to minimise the distress and discomfort to all involved. Nonetheless, we 
recognise that some of the strategies we employed, notably underplaying the 
significance of AIDS to the research, would contravene the injunctions of universal 
 2
ethical codes. Not only did we fail to fully inform our participants of our agenda, but 
we may also have contributed to the dangerous silencing of the pandemic. Through 
problematising individual research encounters, we argue that denial might, with 
certain provisos, be justifiable as part of an ethic of care. We conclude that ethical 
research needs to take account of the emotions of both researchers and researched, 
but also to acknowledge that emotions alone do not present us with clear-cut 
answers. 
 
Emotions and ethics 
Emotions have been neglected, not only by geographers, but throughout the social 
sciences more generally. This neglect of emotion is usually traced to the Cartesian 
separation of mind and body that has characterised Western thought from the 
Enlightenment onwards. The mind/body dualism is paralleled in a distinction between 
reason and emotion, reason being associated with the mind and emotion with the 
body. Mind and reason are accorded a higher status such that ‘[i]n the end, not only 
was emotion not rational, even studying it was probably not rational’ (Damasio 1999, 
p. 39). 
 
The distinction between reason and emotion was particularly promoted by Immanuel 
Kant and extended to the realm of ethics. In Kantian terms, ethics is governed by 
universal principles which are established through the application of reason. 
Emotions are seen as irrelevant: ‘the exclusion of emotions from ethics was based on 
their supposed irrationality (or nonrationality), their physiological, bodily, and ‘bestial’ 
nature, their capriciousness, their apparent involuntariness’ (Solomon 1996, p. 533). 
 
These views have been progressively challenged from a number of quarters. 
Nietzche, for instance, believed life should be governed not by reason but by 
uninhibited emotions (Solomon 1996). Eighteenth century Scottish philosopher, 
David Hume declared that ‘reason is, and ought to be, the slave of the passions’ 
(cited in Solomon 1996, p. 531). Hume developed a ‘moral sentiment theory’ wherein 
morals were based on a set of emotions he termed ‘sympathy’ (Solomon 1996). More 
recent critics, however, have questioned the fundamental dualisms underlying the 
exclusion of emotion from understandings of the ethical domain. 
 
First, some have questioned the extent to which emotion originates in the body and 
not the mind. Solomon (1997), for instance, disputes the dominant view that 
emotions are physiological and essentially irrational aspects of our animal nature. 
 3
Instead, he understands emotions as ideational – a form of cognition rather than 
simply physiological responses.  
 
Second, research has demonstrated the role of emotion in rational thinking. On the 
basis of experiments undertaken with brain-damaged people, Damasio (1999, p. 41) 
concludes: ‘[i]t certainly does not seem true that reason stands to gain from operating 
without the leverage of emotion. On the contrary, emotion probably assists 
reasoning, especially when it comes to personal and social matters involving risk and 
conflict.’ The centrality of emotions to rationality is increasingly argued, particularly in 
relation to the roles of preferences, values and judgement in decision-making 
(Benner 2000; Wenstop and Magnus 2001). ‘[T]he ideas that constitute emotions are, 
at least in part, evaluative judgements’ (Solomon 1997, p. 292). Damasio (1994) 
suggests that even the most primitive perception entails evaluation. As a 
consequence, a growing role is seen for the emotions in ethical thought and action. If 
emotions are judgements, then they are inevitably implicated in ethics (Solomon 
1996). For Damasio (1999, p. 55), ‘[e]motions are inseparable from the idea of 
reward or punishment, of pleasure or pain, of approach or withdrawal, of personal 
advantage and disadvantage. Inevitably, emotions are inseparable from the idea of 
good and evil.’ However, while emotions may be fundamental to consciousness, 
existing prior to conscious reasoning, they cannot be a substitute for reason 
(Damasio 1999). Judgement need not be conscious, reflective or articulate (Solomon 
1997). The contribution of emotion to our judgement of right and wrong may be 
subtler: acting on a perceived moral obligation may, for instance, give the individual a 
sentiment of approval (Solomon 1996). 
 
So long as emotion is seen as a purely individual phenomenon, however, it can only 
relate to a narrow and limited conception of ethics. A deeper imbrication of emotion 
and ethics becomes apparent when emotion is recognised as not only embodied and 
cognitive, but also in part socially constructed. Here, Solomon and Damasio part 
company. Damasio (1999) distinguishes six ‘primary emotions’ (happiness, sadness, 
fear, anger, surprise and disgust), which he ascribes to the individual, from other 
‘secondary’ or ‘social emotions’ (e.g. embarrassment, jealousy, guilt, pride), ascribed 
to the social realm. Solomon, by contrast, while recognising that there are many 
kinds of emotions and that not all can be understood in the same way (Solomon 
1996), argues that ‘most of our emotions are social, not only in their context but in 
their content. They are about other people, even when they seem to be about 
ourselves’ (Solomon 1997, p. 300). Solomon (1997, p. 297) defines emotions as 
 4
‘ways of viewing and engaging the world.’ They are not ‘something inaccessibly 
‘inner’ and ‘private’ … but a public and interpersonal phenomenon’ (Solomon 1997, 
p. 293). Although partly biological-physiological, they are ‘in part learned and 
cultivated in society’ (Solomon 1997, p. 291). Benner (2000), too, asserts that the 
‘primary emotions’ are always about something or in relation to something, and 
hence are social, along with background emotions (what might be termed ‘feelings’, 
e.g. well-being, malaise, calm or tension) which, while not specifically related to 
current events, are nonetheless embodied and impinge on relationships with others. 
 
This view of emotions as social is fundamental to the ‘ethics of care’ popularised by 
Carol Gilligan (1982). Gilligan, a psychologist working within feminist theory, stresses 
emotions such as care and trust as the cornerstone of ethics. In an account of the 
gendering of Western ethics, she states that: ‘[t]he moral imperative that emerges 
repeatedly in interviews with women is an injunction to care … For men, the moral 
imperative appears rather as an injunction to respect the rights of others and thus to 
protect from interference the rights to life and self-fulfilment’ (Gilligan 1982, p. 100). It 
is the Kantian-inspired and masculine-coded ethics of rights that have come to 
dominate Western thinking on ethics. The application of universal principles of 
fairness and means-ends rationality emphasises individuality and assumes an 
autonomous moral agent. In contrast, ‘concern with relationships appears as a 
weakness of women rather than as a human strength’ (Gilligan 1982, p. 17, after 
Miller 1976).  
 
For Gilligan, women’s moral judgements, characterised by inter-personal emotions of 
compassion and tolerance, represent an undervalued ethics of care. Gilligan calls for 
a situational ethics, wherein relationships with others are of paramount importance 
(Benner 2000), calling into question ‘the idea that ethical practice necessarily entails 
the consistent application of universal principles’ (Gormley and Bondi 1999, p. 259). 
 
An ethics of care carries with it a particular conception of moral agency. For Benner 
(2000, p. 17), ‘[o]ur moral agency is embodied, and socially embedded, even in our 
most independent actions.’ An overly cognitive notion of action as embodying choice 
but not emotion or effort is believed to lead to a diminished understanding of agency 
(Campbell 1999). Furthermore, ‘a moral stance in the face of tragedy requires 
compassion for the commitment to be effective’ (Ruiz and Vallejos 1999, p.5). 
 
 5
‘Moral agency, then, can be less mechanistic and less deterministic than 
Descartes’ vision of the subjective mind acting on an object world through 
clear intentions. Also it can be considerably more responsive and generative 
than the Kantian vision or the autonomous choice maker directed by pure will 
and intellect, uninfluenced by emotion and emotional affinities. Both of these 
visions of the person offer a profound respect for the powers of the individual 
to choose or act, but they ignore our embodied interdependence that provides 
the very ground for doing so’ (Benner 2000, p.16). 
 
Care ethics have been challenged on the basis that it is difficult to envisage how 
involuntary emotions can form a basis for ethical action. According to some care 
ethicists, caring is a natural impulse and not fulfilled as a duty. Indeed, for Benner 
(2000), caring fuelled by guilt is often accompanied by contempt for and disinterest in 
the people being helped. Yet if we have no control over our emotions, it would seem 
that can we have no responsibility either. ‘Either you feel sympathy or you don’t; and 
if you don’t, how can the ‘ethics of care’ make any moral demand on you?’ (Paley 
2002, p. 140). This critique is, however, based on an understanding of emotions as 
purely physiological. If emotions are understood as partly cognitive, this allows us 
greater responsibility for the ways we respond to others. For Solomon (1997, p. 297), 
emotions ‘are not (usually) deliberative, and they are often spontaneous, habitual, 
unthinking, ‘natural’ … they are not full-blooded intentional actions, but, more 
important, they do not just happen to us either. They are an essential part of our 
repertoire of responses to the world.’ While we may lack full command of our 
emotional reactions, emotions are part of the way we actively engage with the world, 
and we are not powerless in that engagement.  
 
The view that emotion can partly serve intention may also be expressed as moral 
motivation, ‘the sentiments which induce people to act in pursuit of a more equal and 
caring world’ (Silk 2000, 306). This is central to many religions. Zizek (2000), for 
instance, argues that ‘agape’, the Christian concept of love as charity, should be 
viewed ‘as a self-suppressing duty to love neighbours and care for them, as hard 
work, as something to be accomplished through the strenuous effort of fighting and 
inhibiting one’s pathological inclinations’ (cited in Cloke 2002, p. 594).  
 
The second serious difficulty that many have with the notion of an ethics of care is its 
seeming failure to address questions of equity and justice (Paley 2002). Our ethical 
obligation towards ‘distant strangers’ is rendered problematic, both by rejection of the 
 6
principle of universal fairness, or impartiality, as the primary basis for moral action 
(Smith 1994), and by the demand that moral action should be rooted in specific 
caring relationships with others (Smith 1999). Gilligan’s original work is often unfairly 
accused of rejecting universalist principles, although she actually calls for a 
convergence of ethics of justice with ethics of care: ‘[w]hile an ethic of justice 
proceeds from the premise of equality – that everyone should be treated the same – 
an ethic of care rests on the premise of nonviolence – that no one should be hurt. In 
the representation of maturity, both perspectives converge in the realization that just 
as inequality adversely affects both parties in an unequal relationship, so too violence 
is destructive for everyone involved’ (Gilligan 1982, p. 174). As Damasio (1999, p. 
35) points out, ‘[h]uman emotion is … also about the satisfaction of seeing justice 
served’. 
 
Geographers have debated the merits of a universalist morality compared to a 
situated or local morality (Corbridge 1998; Cutchin 2002; Proctor 1999). Recognition 
is growing that a dualistic approach is unhelpful. Smith (1998, p. 31) criticises, for 
instance, ‘[t]he spatial contention that justice is for the public realm, care for the 
private.’ Whatmore (1997) is equally critical of the simplistic mapping of generalised 
and concrete others onto a global-local binary. ‘The global and the universal are not 
pre-existing empirical qualities; they are deeply fraught, dangerous, and inescapable 
inventions’ (Haraway 1995, p. xix). While emotions may be situated responses to the 
world, situations are not simply local. A range of forms of personal (caring) 
interaction, both mediated and unmediated, take place across physical and cultural 
distance (Corbridge 1998; Robinson 1999; Silk 1998; Smith 1997). 
 
While ethics of care and ethics of justice do not map directly onto local/global 
relationships, the application of an ethics of care to people living in distant places and 
different cultures raises the question of the cultural specificity of emotions. ‘[I]f 
emotions are constituted by ideas, and ideas (at least those ideas that constitute 
emotions) vary from culture to culture, then emotions will vary from culture to culture 
too’ (Solomon 1997, p. 296). Furthermore, there are too few physiological causes 
and distinct sensations to account for the enormous variety of emotions across the 
world (Solomon 1997). ‘Compassion, or some sort of distanced concern for others, 
seems to be the presupposition of every society and it is built into virtually every 
philosophy … But what ‘compassion’ means … is sufficiently different to give us 
pause, and even fear – not to mention different attitudes of fear and pain – is not 
obviously ‘the same’ in all societies’ (Solomon 1997, p. 300). If ethics are related to 
 7
emotions but emotions are culturally specific, it becomes more difficult to argue for a 
universal morality.1
 
The research process: emotions, ethics and codes of practice 
If emotions affect how we, as academics, think and act in relation to other people, 
they are clearly relevant to the conduct of research. Both researcher and researched 
have emotions, and these need to be taken into account. ‘[A]cknowledgeing 
emotions and emotional exchanges orientates us differently within our research 
interviews’ (Laurier and Parr 2000, 99). While Laurier and Parr question whether it is 
appropriate to think in terms of managing emotions in an interview, in practice, 
researchers do take measures to both pre-empt and respond to likely emotional 
reactions. Furthermore, when we conduct research among ‘distant strangers’, the 
interpersonal encounters that we have with them take place within contexts that 
involve both global and local dimensions. We operate in what Mary Louise Pratt 
(1992, p. 7) terms a ‘contact zone’, of ‘copresence, interaction, interlocking 
understandings and practices.’ We need, for instance, to take into account the fact 
that our respondents’ emotional responses to given situations may differ from our 
own and that we may have difficulty understanding them, and they us as a 
consequence. But we need also to recognise that both we and they respond 
emotionally to the global inequalities and power relations that are manifest in the 
research encounter. This is not a simple case of situational versus universal ethics.  
 
The difficulties of justifying ethical research guidelines in practical situations have 
been acknowledged by geographers (Valentine 2003). ‘The imposition of some 
universal code of professional ethics, even within the confines of a national institution 
such as the Association of American Geographers or the Royal Geographical 
Society, faces the opposition of those whose ethics are necessarily contextual’ 
(Smith 1999, p. 283). Nonetheless, many researchers subscribe to codes of research 
ethics to justify the ethical soundness of their research. One characteristic of these 
codes, and the focus of this paper, is an insistence on openness about the research. 
The need for informed consent, in which full details of the purpose of the research 
are made available to participants, is stressed in most recent guidelines concerning 
research. The British Sociological Association (2002), for instance, states: ‘As far as 
possible participation in sociological research should be based on the freely given 
informed consent of those studied. This implies a responsibility on the sociologist to 
explain in appropriate detail, and in terms meaningful to participants, what the 
research is about …’ Most guidelines for research with children advocate that 
 8
children are fully involved in all aspects of the research from planning to 
dissemination.2 As Cree et al (2002, p. 48) point out, however, these codes ‘offer 
topics for consideration rather than ‘blue-prints’ for good practice’. Geographers have 
argued that they should not be bound by the requirement for informed consent when 
researching the lives of people who cannot give such consent due to profound 
intellectual disabilities (Metzel 2000), or where respondents are opposed to the social 
justice agenda of the researcher (Wilton 2000), provided the respondents are not 
harmed by the research. 
 
In the case study below, we discuss research in which we were not entirely open to 
participants about our purpose. The reasons for this relate to the way in which 
emotions would have entered into the research encounter had we been explicit about 
our purpose. We argue that we were applying a situated ethics that, rather than 
abandoning concern for global justice, was prompted by emotional responses to, 
among other things, the absence of justice, although we cannot say with total 
confidence that we ‘did no harm’. 
 
Case study: Young AIDS migrants research 
In 2001 we conducted research among urban and rural communities in Lesotho and 
Malawi with the purpose of investigating the experiences of young people who had to 
move house as a consequence of AIDS among their relatives. These included 
children whose parents/ guardians had become sick or had died and were no longer 
able to care for them, and others who moved because their help was needed 
elsewhere (see Ansell and van Blerk 2004; Young and Ansell 2003). In the course of 
the research we asked children about the circumstances of their relatives’ 
illnesses/deaths, about how they had felt and the impacts of the death/sickness upon 
them. When interviewing guardians, we again asked about causes of illness and 
death, and about the impacts on the children concerned and on other household 
members. Rates of HIV infection in both Lesotho and Malawi are very high, and 
about two thirds of orphans in both countries are orphaned by AIDS 
(USAID/UNICEF/UNAIDS 2002). Nonetheless, although we asked people about 
sickness and death among their relatives we were less candid in revealing that our 
research agenda was focused on the impacts of HIV/AIDS. Such an approach is not 
uncommon: others researching the social impacts of AIDS in Africa also use a proxy 
such as chronic illness, particularly where, as was the case for us, the cause of 
illness is seldom closely related to its consequences. 
 
 9
Contextualising the research encounter: AIDS and its effects 
Across sub-Saharan Africa 7.5% of 15-49 year old adults are living with HIV; in 
Swaziland, Botswana and Lesotho numbers exceed 25% (UNAIDS 2004). This is 
only a snapshot picture: an extreme projection suggests that, if current trends 
continue, as many as 90% of 15-year-old boys in Botswana will ultimately die of 
AIDS (UNAIDS 2000). For the vast majority of individuals, a positive diagnosis 
means inevitable death within a few years. Few Africans survive more than five years 
post-diagnosis (Schoepf 2001). AIDS is not merely a death sentence, but also has 
negative consequences for a person’s remaining life, and that of their families. AIDS 
causes very unpleasant illnesses which are costly to treat. People living with 
HIV/AIDS (PLWHAs) suffer not only illness, but increasing poverty, associated with 
costs of treatment and inability to work. Where a productive person becomes sick, 
their income or value of their labour is lost to their family, as is that of those whose 
time must be devoted to caring for them. AIDS-affected households commonly 
become impoverished. Furthermore, the children of those who become sick or die of 
AIDS may be left with no property and inadequate arrangements for their care. A 
diagnosis (or even suspicion) of AIDS brings fear and despair, not only for the 
individual directly affected but also their family.  
 
Emotions are produced in social contexts that are interpreted (in part through 
emotion) in relation to both universalised ethics of justice and more localised 
moralities. The nature of the AIDS pandemic in southern Africa is such that research 
encounters with AIDS-affected people need to take into account the emotions that 
arise from both the immediate personal situations of the participants but also wider 
contexts of injustice. While writing on emotion and ethics has focused on the positive 
emotions that ground ethics of care, research with AIDS-affected people has also to 
address more negative feelings of anger, guilt, shame and discomfort. Four inter-
related framing contexts are considered below, each of which is shown to have 
shaped the intersubjective production and expression of emotion within the Young 
AIDS migrants research. 
 
1. Global inequalities 
That AIDS is wreaking disaster on African individuals and families, when Africa is 
already so impoverished, is likely to confront most observers as unjust: a sense of 
injustice that is both rationalised and emotionally felt. While global AIDS discourse 
focuses on individual risk behaviour, risk, in reality, stems not so much from 
ignorance of AIDS as from the precarious situations in which millions of people live 
 10
(Farmer et al 2001). Such situations have worsened in recent decades as a 
consequence of international economic regimes (Schoepf 2001). Both globally and 
locally, ‘the disease is distributed unequally across populations in line with pre-
existing socio-economic inequalities that could be considered immoral and unjust’ 
(Ashforth 2001, p. 8). The differential availability of care and treatment resources in 
different parts of the world appears even more unjust when seen in the context of a 
universal ethics (Baylies and Bujra 1997). Over the past few years the global 
differences in prognosis for those infected with HIV have become much sharper. New 
treatments, known collectively as ‘Highly Active Antiretroviral Therapies’ (HAART), 
‘have the capacity to transform AIDS from an automatic death sentence to something 
closer akin to a chronic illness’ (Baylies and Bujra 1997, p. 385). Yet these 
treatments, which now cost about $200 per person per year, and would not only save 
lives, but also benefit families and societies more widely, are unavailable to the vast 
majority of PLWHAs across sub-Saharan Africa, even while the US government 
budgets about $45 million a year to supply its troops with Viagra (Silverstein 1999).  
 
This global injustice frames the production and expression of emotion in any 
research encounter with AIDS-affected people in sub-Saharan Africa. Knowledge 
that the opportunity for life available to people in many Western countries is 
unavailable to PLWHAs in Africa adds to the sense of guilt that is so often a feature 
of research in Third World societies. This knowledge, or at least awareness of their 
own poverty and our relative wealth, also colours the reactions of research 
respondents. Farmer (1994) drew attention to the way Haitian people associate AIDS 
with wider political-economic processes, including North American imperialism. Many 
people we encountered in the course of our research asked for material help. 
However much we believe our research to be of value to wider southern African 
society, the fact that our research budget did not provide for us to meet the 
expectations of the individuals we met generated a sense of impotence (as is often 
the case when conducting research across such material inequality).  
 
In conducting research concerning AIDS we encountered not only people who were 
upset and sorrowful, but also, occasionally, people who were angry. In Thibella, a 
particularly poor neighbourhood of Maseru, people became quite hostile towards us, 
because we promised no material support – possibly because our visit coincided with 
that of an NGO AIDS project which was well resourced. Here we felt that research 
participants deliberately played on our emotions – people would tell tragic stories in 
the hope of receiving assistance; one woman’s tears, which began before the 
 11
research was even introduced to her, appeared to be designed to evoke our 
sympathy and guilt. In practice, the emotions aroused were more complex – 
annoyance that people were so demanding, manipulative and unwilling to cooperate, 
but also guilt at feeling annoyed when people clearly are in need. The emotional 
reactions of both researchers and researched to global inequalities that exacerbate 
the damage caused by AIDS thus inhibit free dialogue on the subject.  
 
2. Racialised discourse 
Our research relations with AIDS-affected southern Africans are structured not only 
by global economic inequalities, but also by the racialised nature of global (and local) 
AIDS discourse, with its roots in European colonialism. Paradoxically, AIDS carries 
associations both with whites and with Africans. Racism and xenophobia were 
apparent in explanations of the presumed origins of the epidemic, where images 
were invoked of sex between humans and animals, or eating the raw flesh of green 
monkeys (Aggleton and Parker 2002). This racism has both imbued dominant 
images of the epidemic, and been reproduced through them (Aggleton and Parker 
2002). Sub-Saharan Africa is the global region with by far the most severe 
experience of AIDS. Biomedical discourse has cast the blame for AIDS on individual 
risk behaviour which in Africa has been attributed to cultural difference (Schoepf 
2001). Given the sexually transmitted nature of the disease, inferences are drawn by 
some about African sexual practices (although there is no demonstrable link between 
HIV prevalence and cultural differences in sexual behaviour (UNAIDS 2002)). 
Africans are thereby portrayed as the causes of their own misfortunes (Aggleton and 
Parker 2002).  
 
At the same time, AIDS is often associated by African people with outsiders - it is 
said to come from ‘elsewhere’. There is a perception among many in Nigeria that 
AIDS is a disease of foreigners (Alubo et al 2002). Sometimes, particularly in 
southern Africa, AIDS is associated specifically with whites. Both the ‘indigenous’ 
white population and visitors are blamed for introducing a white ‘gay plague’ into 
Africa’s heterosexual population. In Lesotho, the first reported case of AIDS was a 
white man, a fact that dominated people’s perceptions of the disease for some years. 
‘In the eyes of some African and Asian leaders, HIV/AIDS has been viewed as a 
disease of the West, linked to the weakness of family structures, liberal social values 
and moral decline’ (Aggleton and Parker 2002, p. 9). 
 
 12
These harmful racialised discourses have roots in and perpetuate colonial 
relationships between Africa and the West. They are present in the research 
encounter, but unlike economic injustices remain unspoken. They are felt as 
discomfort: as white women, we feel conscious of our bodies and uncomfortable in 
our skins. Emotions are aroused in researchers and researched, but all are reluctant 
to speak in such a way as to bring these to the foreground. These discourses, then, 
contribute to the difficulty of an open dialogue about AIDS. 
 
3. Stigma 
Research with AIDS-affected people must also take into account the stigma that 
informs emotional impacts of, and reactions to, the disease. As early as 1987, 
Jonathan Mann, then director of the WHO Global Programme on AIDS, described 
three phases of the epidemic: the epidemic of HIV, the epidemic of AIDS, and the 
epidemic of stigma, discrimination and denial. In his assessment, the latter was ‘as 
central to the global AIDS challenge as the disease itself’ (Mann 1987, cited in Parker 
and Aggleton 2002). While AIDS is almost universally stigmatised, this plays out 
locally in different ways. 
 
Stigma exists in relation to moral codes. As ‘a construction of deviation from some 
ideal or expectation’ (Alonzo and Reynolds 1995, cited in Taylor 2001), stigma is an 
instrument of social control, expressed through disidentification, with stigma 
commonly experienced by individuals as shame (rather than guilt, which is a form of 
self-control (Solomon 1996)). The attachment of stigma to disease is not uncommon, 
particularly where the disease is severe, disfiguring, incurable and progressive 
(ICRW 2002a). Such conditions evoke strong emotional responses such as fear and 
revulsion (Taylor 2001). The stigmatisation is more marked where disease 
transmission is perceived to be a result of individual behaviour, especially where this 
transgresses social norms (ICRW 2002a). Thus AIDS-related stigma interacts with 
pre-existing prejudices connected with sexuality, gender, race and poverty (Parker 
and Aggleton 2002). Fears about contagion and disease are thereby compounded by 
deep-rooted anxieties about social breakdown. 
 
According to UNAIDS (2001, p. 1), ‘stigma … undermines prevention, care and 
support; it also increases the impact of the epidemic on individuals, families, 
communities and nations.’ Stigmatisation ‘silences and saps the strength of already-
weakened individuals and communities, and causes people to blame themselves for 
their predicament’ (Aggleton and Parker 2002, p. 9-10). Stigma has psychological 
 13
consequences, including depression, lack of self-worth, despair (Aggleton and Parker 
2002), and since the immune system of a PLWHA is physically affected by their 
emotions (Weiser 1999), stigma and shame can damage an individual’s physical 
health. Moreover, secondary stigma may become attached to the families and friends 
of those infected (ICRW 2002b). In societies where cultural systems emphasise 
collectivism over individuality, HIV/AIDS may be perceived as bringing shame on the 
family and community (Parker and Aggleton 2002).  
 
Stigma is harmful, both in itself, and because it leads to discrimination against those 
stigmatised (Aggleton and Parker 2002). It arouses emotions of not just shame 
among those with HIV/AIDS, but also fear. People in Tanzania, for instance, 
expressed more concern about rejection and stigma than about the technical facts of 
AIDS (Lie and Biswalo 1994). Unsurprisingly, given the potential consequences, 
many people avoid finding out about their HIV status. Most PLWHAs who know their 
HIV status deny or hide their condition (Alubo et al 2002), and even continue to 
practise unsafe sex for fear of arousing suspicion (Aggleton and Parker 2002), 
putting themselves and others at further risk.   
 
There is a need for anyone conducting research with AIDS-affected people to be 
sensitive to the stigmatisation of the disease, and to avoid any possibility that those 
involved in the research as participants might suffer stigma as a consequence. Some 
research in the past has arguably contributed to the stigmatisation of AIDS, 
particularly where it has represented PLWHAs as vectors of transmission rather than 
as people (Baggaley and van Praag 2000), and associated the disease with ‘risk 
groups’, inevitably stigmatising group members (Goldin 1994), and emphasised 
individual behaviour change, thereby casting responsibility on those who lack the 
power and autonomy to control their lifestyles.  
 
Stigma shapes both the production and expression of emotion in research. The 
effects of stigma on research encounters are most likely to be manifested in silence. 
If a person affected by AIDS fears stigma, they are unlikely to be comfortable 
discussing AIDS in a research encounter. If researchers are sensitive to such fears, 
speaking of AIDS may be very difficult. The researchers’ own concerns (perhaps 
culturally determined) about causing respondents to feel shame may paralyse any 
possibility of discussion.  
 
 14
4. Immediate circumstances 
Not only global power relations and societal moralities impinge on the research 
encounter such as to provoke emotional responses, but researchers intrude into the 
local worlds of their respondents which may be charged with emotion. The situations 
that confront individual AIDS-affected people differ in their details, but are frequently 
harrowing. 
 
Matseliso, for instance, was living in Maseru, caring for her own two children and two 
of her dead sister’s children when she was interviewed. She also supported her older 
nieces and nephews who now lived alone, their parents having died. On the wall was 
a wedding photograph, in which she and her husband conformed to the Basotho 
ideal of plumpness. Her husband, a soldier, had died a year ago, and Matseliso was 
now very thin and clearly close to death. Yet in the course of the interview she 
constantly talked about getting better, perhaps because her child was present. 
 
Talking with people about experiences associated with sickness and/or death among 
close relatives can provoke distress. A small number of the people we interviewed, 
both adults and children, were reduced to tears when talking about their own 
situations, especially about relatives who had died. Being in a situation where people 
we know little are in a highly emotional state, particularly where they are distressed, 
causes discomfort for the interviewer. Researchers are therefore likely to avoid or 
move discussion away from topics that arouse such emotions among interviewees. 
 
Managing emotions in research 
In situations where research encounters are likely to give rise to heightened 
emotional responses, it is appropriate to attempt to manage the research such that 
the emotions that are provoked harm neither the research participants nor the 
researchers nor the research itself. Assessing what is harmful is not easy, but below 
we outline three ways in which we sought to manage the way emotions intervened in 
the young AIDS migrants research outlined above. We then move on to consider how 
such strategies sit ambiguously in relation to guidelines for ethical research. 
 
1. Dealing with distress 
We did not desist from talking about sickness and death with children and adults, 
despite our awareness that such conversations may be distressing for some 
participants. On the very few occasions that a person clearly became upset, the 
interview was ended, unless, as usually happened, the interviewee expressed a wish 
 15
to keep going. One girl in Lesotho whose mother had recently died became very 
upset and cried, but wanted to continue talking about her situation. It is perhaps 
significant to recognise that as researchers we may not be causing the distress, but 
merely provoking it into the open, and thereby exposing ourselves to it. While this is 
uncomfortable for the researcher, the interviewee is not necessarily ‘harmed’ by the 
experience. Distress associated with sickness and death is ‘natural’; talking and 
releasing emotions, whether of sadness or anger, can be cathartic. This is not to 
suggest that it is appropriate to subject a person to serious and prolonged distress 
(see Robson 2001). We would argue, on the basis of ethics of care, that the way in 
which we respond to such a situation should be consistent with both our emotional 
responses to the situation and theirs.  
 
Our ability to deal with situations in which people become distressed is, however, 
inhibited by the fact that, as researchers, we do not expect to deal with people’s 
emotions. Because the role of emotions in research is ignored, no attention has been 
given to training researchers in dealing with such situations. As a consequence we 
cannot feel confident in dealing with people in distress without causing harm. 
Moreover, in the research reported here, we generally had only fleeting contact with 
the people whose lives we were researching, and were unable to communicate 
directly with people in their own languages. We were fortunate in that we were 
supported in some of the research by people who were trained in counseling. Our 
interpreter in Lesotho, who assisted with all interviews, was a counselor, and in 
Malawi the interviews with adult guardians and with street children were undertaken 
in the presence of trained counselors, working for local NGOs. Most other children 
were interviewed in school settings, with the support of their teachers close at hand.  
 
2. Permitting denial 
Although we were funded to conduct research specifically concerning young people's 
AIDS-related migration, the research agenda we presented to school children and 
adult community members focused on children’s migration. Although we asked 
people about sickness and death in relation to their own families, and responded to 
emotional distress if and when it occurred, we did not ask anyone directly about 
AIDS. For the reasons outlined earlier, we felt that discussion of AIDS could arouse 
emotional responses that would be difficult for both the research participants and 
ourselves, and may hinder the progress of the research. 
  
 16
In conjunction with our own silencing of our research agenda, we allowed participants 
to deny AIDS. If a family member had died, causing a child to move home, we 
enquired as to the cause of death. Many children, in particular, claimed they did not 
know what their parents had died of - they may not have known, or may have sought 
to hide a (suspected) truth. Many people, both adults and children, talked of 
symptoms rather than specifying diseases (see Table 1). No one named AIDS, 
though two people alluded to it as ‘the youngsters’ disease’ or ‘the common disease’. 
‘Something in the chest’ and ‘ulcers’ are reportedly common euphemisms for AIDS in 
Lesotho. Many of the symptoms and diseases mentioned (most notably tuberculosis) 
could well be AIDS-related. Others (stitch, a slight headache) are clearly not normally 
fatal, and perhaps represent denial of AIDS. If a person told us their relative had died 
of a pain in the chest, or of stitch, we did not challenge them or enquire whether it 
might have been AIDS-related. When deaths were attributed to causes we suspected 
might be AIDS, we could not know whether our informants were aware (or 
suspicious) of the true cause of death, whether they wanted to keep it from us, or 
whether they simply could not speak of it.  
 
Table 1: Explanations offered for deaths of adult relatives 
Lesotho    Malawi    
TB 20 cancer 1 TB 15 headache and 
nose bleeds 
1 
something in 
chest/chest pain 
6 heart attack 1 malaria 8 slight headache 1 
car accident   5 bewitched 1 headache(s) 6 slight 
headache/minor 
infection 
1 
pneumonia 3 diabetes 1 sick 6 headache/fever/i
nfection 
1 
shot 3 deep cold 1 bewitched 4 diarrhoea and 
headaches 
1 
stabbed 2 ulcers 1 diarrhoea 3 lump in neck 1 
‘murdered’   2 arthritis 1 ‘a disease’ 3 lump under arm 1 
stroke 2 lung disease  1 problems with 
legs 
2 paralysis of legs 1 
stitch 2 beaten to death 
in conflict 
1 ‘the common 
disease’ 
2 stomach/bowel 
problems 
1 
childbirth 2 stress at 
another’s death 
1 childbirth 2 stomach ache 1 
growth on brain 1 ‘things coming 
out of vagina’ 
1 pains/swelling 1 swellings in 
stomach 
1 
multiple disease 1 ‘youngsters’ 
disease’ 
1 pains and 
headache 
1 typhoid 1 
something 
wrong with 
bones 
1 lungs full of 
water 
1 swelling 1 cancer 1 
    swollen leg 1 stroke 1 
    shingles 1 fell down well 1 
    cough 1 murdered 1 
 17
    malaria/heart 1 car accident 1 
    headache -> 
madness 
1 heart 
failure/blood 
pressure 
1 
Source: focus groups with children; interviews with adult guardians 
 
We did raise the subject of AIDS, but not in such a way as to imply that we suspected 
that those whose sickness or deaths we were discussing might have HIV. We asked 
children what they knew of AIDS, how it was contracted, and whether anyone in their 
communities or families was affected by it. All children denied knowing anyone 
affected, and most reported that those affected by AIDS were ‘people with 
unbecoming behaviour’ or ‘dirty people’, revealing the extent to which AIDS remains 
stigmatised. In interviews with adults, we asked about deaths in their communities 
and whether they knew the cause of these deaths, and also about whether the 
growing number of orphans resulted from AIDS. Although in Malawi, some people 
attributed deaths in their communities to AIDS, in Lesotho they did not.  
 
By allowing participants to remain silent about AIDS within their families, we felt that 
we were protecting them from an interpersonal situation that could be emotionally 
taxing. We were doubtless protecting ourselves from discomfort, too. Furthermore, 
we were protecting the research: had we revealed at the outset that we were 
interesting in talking to people whose relatives had become sick or died of AIDS, it is 
extremely unlikely that anyone would have been willing to speak with us. The actual 
cause of death was not of great importance to the research as the consequences for 
children of sickness and death differ little irrespective of the cause.  
 
3. Tailoring dissemination  
We have disseminated our research findings in several ways in the countries where 
the research was conducted. Before leaving, we made posters for the participating 
schools. These reported the preliminary findings from the migration questionnaire 
results and did not mention AIDS. We also conducted workshops with NGO and 
government representatives. Here we related our research directly to AIDS. We 
subsequently produced two different versions of the final report. One report was 
distributed to schools and local communities. This was carefully edited and given the 
title ‘Young Migrants in Southern Africa: moving in the wake of AIDS’ so as not to 
imply that the children and families we had worked with were themselves affected by 
AIDS, as this could have stigmatised them within their communities. The relevance of 
the research in relation to the high incidence of AIDS was, however, discussed. By 
 18
contrast, the ‘official’ report, entitled ‘Young AIDS migrants’, along with academic 
papers we have produced, explained that although we did not specifically ask about 
AIDS, we could legitimately infer that about two-thirds of the orphans in our study 
were in fact AIDS orphans. 
 
Our caution concerning dissemination to research participants was motivated by a 
concern not to distress or stigmatise them: protecting ourselves or the research was 
not an issue in this regard. When addressing government and NGOs, however, we 
felt that it was important to make the link with AIDS. Although all orphans and 
vulnerable children may have similar experiences, the research is of greatest 
relevance to practitioners and policy makers when seen in relation to the rapidly 
increasing numbers of such children that is a direct outcome of the AIDS pandemic. 
Even in this context, however, the issue was emotive. In the first dissemination 
phase, some representatives of government and NGOs expressed strong feelings 
that we were wrong to relate our research to AIDS, a response that can best be 
interpreted in light of the global discourses that frame discussions between people 
from the West and from southern Africa on the subject of AIDS.  
 
A sense of right and wrong? 
To a large extent, in our research we did what we ‘sensed’ to be right. The term 
‘sense’ (perhaps too vague to be useful?) suggests both an embodied feeling and the 
application of rationality. We tried to be sensitive to the emotions of others – to their 
‘sensibilities’. Talking about sickness and death, but not about AIDS, both seemed to 
us to be sensitive to those with whom we talked. The application of ‘reason’, and a 
universalist ethic, however, raises two ways in which our avoidance of talking about 
AIDS might be criticised.  
 
1. Breaching ethical codes 
It is clear that in conducting the research we failed to fully inform participants of what 
the research was ‘about’. We concealed from them the fact that we would use what 
they told us to make inferences concerning the impacts of AIDS. As our focus was on 
the indirect impacts of AIDS – impacts which are similar for most orphans and 
vulnerable children, we could have undertaken a similar research project that did not 
have AIDS in its title. This would not, however, have resolved the question of how far 
we should make participants aware of the possible ways we (or others) might 
interpret what we learn in research encounters. 
 
 19
Numerous authors have pointed to the inadequacies of ethical codes as universal 
prescriptions for research practice. Cloke (2002, p. 591) attributes any progress 
towards a more ethically sensitive geography, not to ‘the imposition of new ethical 
codes or guidelines but more because of the introduction of more flexible prompts for 
moral contemplation which have stimulated and nurtured moral imaginations in 
human geography research’. Cree et al (2002, p. 54) suggest similarly, ‘[t]he codes of 
ethics and guidelines for good practice which have been developed can never be 
more than this, because there will always be ambiguities and complexities’, while 
Hay and Foley (1998, 180) go further to argue that ‘[t]he routinisation of ethical 
decision making through rigid institutional codes and practices denies or curbs the 
possibility of responsible citizenship.’  
 
Nonetheless, the sense of having contravened ethical guidelines makes us 
uncomfortable. Wiener (2001) makes a relevant point in relation to the difficulties 
experienced by analysts in reconciling moral principles (codes of ethics) with their 
internal personal ethical attitude where there are dilemmas concerning confidentiality. 
She suggests that most analysts try to find a compromise position: a ‘mental and 
emotional ethical space’ (p.431). 
 
2. Reinforcing stigma 
That stigma is damaging has already been stated. One of the most harmful impacts 
of the stigmatisation of AIDS is the silence it provokes. ‘Fears of the consequences of 
open discourse and self-identification have created a silence that threatens all of us’ 
(Goldin 1994, p.1359). The silencing of AIDS is not only harmful to those who are at 
risk due to lack of knowledge. It has also ‘caused those infected with HIV and 
affected by the disease to feel guilty and ashamed, unable to express their views and 
fearful that they will not be taken seriously’ (Aggleton and Parker 2002, p. 5), and 
even causes people to deny themselves the possibility of treatment, care and 
community support (Baylies and Bujra 1997). The WHO has long recognised denial 
as one of the greatest obstacles to AIDS prevention (WHO Global Programme on 
AIDS 1991), and the silence of political leaders has been particularly damaging in 
this regard (Aggleton and Parker 2002).  
 
The advantages to society in general of more open discussion of AIDS seem clear. 
Yet, despite the difficulties stigma presents for the prevention and treatment of AIDS, 
in our research we operated within the ‘culture of silence’ that surrounds the 
 20
epidemic. In this respect our silence seems unethical – could we not have confronted 
stigma through the research process, by encouraging open dialogue? 
 
Although openness is doubtless ‘good for society’, it does put the individual at risk. 
Goldin (1994) points to the tension between the rights of the directly affected 
individual and those of the rest of society. Research by the ICRW (2002b) in three 
African countries found that while most people expressed a belief that PLWHAs 
should disclose their status, in practice very few felt able to do so, because they 
feared stigma. Given an appropriate context, it might be possible to support people in 
open dialogue. Research among Aboriginal people in Canada suggests that giving 
PLWHAs an environment in which they feel a sense of support and an opportunity to 
express themselves without stigma or judgement encourages them to take better 
care of themselves (Weiser 1999). It has been argued that the concern of the 
Western human rights lobby over confidentiality adds a new layer of AIDS-related 
stigma in African societies (Allen and Heald 2004). Nonetheless, for the reasons 
outlined earlier, an encounter between Western researchers on relatively short visits 
and southern Africans is not an ideal context for breaking down stigma. 
 
An uncomfortable silence? 
Talking openly about AIDS with AIDS-affected people in southern Africa is difficult 
and may, in some circumstances, be impossible. This raises the question of the 
extent to which it is important for people specifically to name AIDS. ICRW (2002b, p. 
3) researchers observed that in interviews people were unwilling to name AIDS 
openly, preferring to talk in abstract terms, or using such descriptors as ‘that disease 
we learned about.’ Elsewhere, people employ a witchcraft metaphor in discussions of 
AIDS. This may not conflict with scientific knowledge of the disease, as Ashforth 
(2001) points out in relation to South Africa. Farmer (1994), similarly, discusses the 
way in which Haitians conceive of AIDS as (among other things) sent through 
sorcery. People would see such explanations as in no way inconsistent with a 
knowledge that it was caused by a ‘microbe’. Metaphors do not necessarily indicate 
ignorance, or even complete denial – just an unwillingness to talk openly. 
Appropriating the diagnosis of witchcraft may even be helpful insofar as it allows 
those affected to lay blame rather than accept blame, and thereby to avoid stigma 
(Baylies and Bujra 1997).  
 
Research in ‘contact zones’ is uncomfortable: the rules that one side finds comforting 
may cause discomfort to the other. As Besio (2003, p. 27) remarks, ‘[m]any, if not 
 21
most, researchers would rather not locate themselves or their research in a contact 
zone, given Pratt’s description [cited above]. Coercion? Radical inequality? 
Intractable conflict? These words describe the antithesis of desirable feminist and 
most if not all research interactions’. It is not particularly helpful, however, to insist 
that those we research with must speak in our language, use our terms: that if we 
must speak openly, not in metaphor, then we cannot communicate at all.  
 
To cease to conduct research where it is impossible to be explicitly open about the 
research without causing a degree of emotional distress to participants would inhibit 
research relating to many areas of human experience. If we are to continue to 
conduct research on such issues it is necessary to make judgements about how 
open to be about the way the research might be interpreted and how cautious to be 
about evoking negative emotions. In making such judgements, researchers should 
be governed by rationality and emotion, or, simply, ‘[t]o act ethically requires a 
person to listen to both head and heart’ (Silk 2000, 306). 
 
Lastly, there is a temporality to the relationship between emotions and research that 
is worth noting. Contexts change over time, and with them the emotions that are 
provoked. Malawi has a longer history of AIDS than Lesotho, and at the time of the 
research, participants discussed the disease more freely there. Although the initial 
research was limited to two and a half months in each country, we returned for a 
further phase of dissemination two years on. People in all four communities 
welcomed our research findings and were by this time much more willing to discuss 
AIDS in relation to its impacts on their own communities. This raises the question of 
whether we should have waited until we could undertake the research in an entirely 
transparent way. On our return to southern Africa, people in the local research 
communities as well as policy makers and practitioners had ‘woken up’ to the 
pandemic and were eager to learn what we had found. We were able to supply a 
knowledge gap that had arisen in our absence, and at this stage to discuss openly 
the impacts of AIDS. Had we not undertaken the research, this would have been 
impossible. Furthermore, it is possible that through our research, particularly by 
raising the subject of AIDS in a non-threatening way, we had not simply generated 
findings to fill a hunger that would arise spontaneously, but had contributed to a 
growing awareness of the pandemic and its existing and potential effects. Rather 
than reinforcing the silencing of the pandemic, our cautious approach may have gone 
a little way towards exposing it. 
 
 22
Conclusions 
Ethical research needs to take account of the emotions of both researchers and 
researched. This is not to suggest a simplistic avoidance of all ‘negative’ emotions on 
either side. We should not be diverted from what we ‘sense’ to be right because of 
our own fear or guilt or embarrassment. Neither should we assume that the 
expression of negative emotions on the part of those we research is necessarily 
harmful (any more than we should assume that arousing ‘positive’ emotions means 
that we are doing good). Our own emotions and the way they make us sensitive to 
the emotions of others are a valuable part of the way we relate with people affected 
by the frightening and distressing impacts of AIDS. 
 
Our emotions do not present us with clear-cut answers. Sense of right and wrong is 
more than just a feeling, it needs to be rationally justifiable. It also confronts us with 
uncomfortable contradictions. Universalist ethics of justice, as expressed in ethical 
codes, might indicate that our research was ethically wrong. Two forms of harm may 
have been implicated: our failure to fully disclose our purpose arguably both deceived 
participants and, perhaps more seriously, contributed to the silencing and related 
stigmatisation of AIDS. However, from the viewpoint of an ethic of care, taking 
account of the emotional context of interpersonal relations, it might be argued that we 
undertook the research in such a way as to avoid certain types of harm: not only 
emotional distress, but also the direct stigmatisation of individuals and families. 
Beyond these varied forms of harm, we should also consider the harm of not having 
undertaken research which was ultimately of interest to participants and potentially 
beneficial, not only for the children involved in our own research, but also others 
similarly affected by HIV/AIDS. Whereas, as academic researchers, we are ill-
equipped to break down stigma and denial at a local level with confidence that we 
are not doing harm, we are able to deliver our findings to, and work with 
representatives of governments and NGOs, who might be able to implement positive 
change locally. 
 
The choices that faced us in undertaking the research were not simply questions of 
global justice versus immediate circumstance or ends versus means, but involved a 
more complex navigation of the intersections of global and local, justice and care. 
‘Without emotion, without caring, a theory of justice is just another numbers game … 
Reason and emotion are not two conflicting and antagonistic aspects of the soul. 
Together they provide justice, which is neither dispassionate nor ‘merely emotional’’ 
(Solomon 1995 cited in Smith 1998, p. 34). Professional and substantive ethics are 
 23
inevitably connected in geography – ‘process and product, context and content, are 
not comprehensible outside of the other’ (Proctor 1999, p. 6). 
 
References 
Aggleton P and Parker R 2002 A conceptual framework and basis for action: 
HIV/AIDS stigma and discrimination UNAIDS, Geneva 
Alderson P 1995 Listening to children: children and social research ethics 
Barnardo's, London 
Allen T and Heald S 2004 'HIV/AIDS policy in Africa: what has worked in Uganda and 
what has failed in Botswana?' Journal of International Development 16, 
1141-1154  
Alubo O, Zwandor A, Jolayemi T and Omudu E 2002 'Acceptance and stigmatization 
of PLWA in Nigeria' AIDS Care 14(1), 117-126  
Ansell N and van Blerk L 2004 'Children's migration as a household/family strategy: 
coping with AIDS in Malawi and Lesotho' Journal of Southern African 
Studies 30(3), 673-690  
Ashforth A 2001 AIDS, witchcraft and the problem of power in post-Apartheid 
South Africa School of Social Science, Institute for Advanced Study, 
Princeton NJ 
Baggaley R and van Praag E 2000 'Antiretroviral interventions to reduce mother-to-
child transmission of human immunodeficiency virus: challenges for health 
systems, communities and society' Bulletin of the World Health 
Organization 78(8), 1036-1044  
Baylies C and Bujra J 1997 'Social science research on AIDS in Africa: questions of 
content, methodology and ethics' Review of African Political Economy 73, 
380-387  
Benner P 2000 'The roles of embodiment, emotion and lifeworld for rationality and 
agency in nursing practice' Nursing Philosophy 1, 5-19  
Besio K 2003 'Steppin' in it: postcoloniality in northern Pakistan' Area 35(1), 24-33  
British Sociological Association 2002 'Statement of ethical practice' , accessed 
21/05/03, http://www.britsoc.org.uk/about/ethic.htm
Campbell C 1999 'Action as will-power' The Sociological Review, 48-61  
Cloke P 2002 'Deliver us from evil? Prospects for living ethically and acting politically 
in human geography' Progress in Human Geography 26(5), 587-604  
Corbridge S 1998 'Development ethics: distance, difference, plausibility' Ethics, 
Place and Environment 1(1), 35-53  
 24
Cree V E, Kay H and Tisdall K 2002 'Research with children: sharing the dilemmas' 
Child and Family Social Work 7, 47-56  
Cutchin M P 2002 'Ethics and geography: continuity and emerging syntheses' 
Progress in Human Geography 26(5), 656-664  
Damasio A 1999 The feeling of what happens: body, emotion and the making of 
consciousness Heinemann, London 
Farmer P 1994 'AIDS-talk and the constitution of cultural models' Social Science 
and Medicine 38(6), 801-809  
Farmer P et al 2001 'Community-based approaches to HIV treatment in resource-
poor settings' The Lancet 358, 404-409  
Gilligan C 1982 In a different voice: psychological theory and women's 
development Harvard University Press, London 
Goldin C S 1994 'Stigmatization and AIDS: critical issues in public health' Social 
Science and Medicine 39(9), 1359-1366  
Gormley N and Bondi L 1999 'Ethical issues in practical contexts' in Proctor J D and 
Smith D M (eds) Geography and ethics: journeys in a moral terrain 
Routledge, London 
Haraway D 1995 'Cyborgs and symbionts: living together in the new world order' in 
Gray C H (ed) The cyborg handbook Routledge, London 
Hay I and Foley P 1998 'Ethics, geography and responsible citizenship' Journal of 
Geography in Higher Education 22(2), 169-183  
ICRW 2002a Addressing HIV-related stigma and resulting discrimination in 
Africa: a three-country study in Ethiopia, Tanzania and Zambia 
International Center for Research on Women, Washington DC 
ICRW 2002b Understanding HIV-related stigma and resulting discrimination in 
sub-Saharan Africa: emerging themes from early data collection in 
Ethiopia, Tanzania and Zambia International Center for Research on 
Women, Washington DC 
Laurier E and Parr H 2000 'Emotions and interviewing in health and disability 
research' Ethics, Place and Environment 3(1), 98-102  
Lie G and Biswalo P 1994 'Perceptions of the appropriate HIV/AIDS counsellor in 
Arusha and Kilimanjaro regions of Tanzania: implications for hospital 
counselling' AIDS Care 6(2), 139-151  
Metzel D 2000 'Research with the mentally incompetent: the dilemma of informed 
consent' Ethics, Place and Environment 3(1), 87-90  
Paley J 2002 'Virtues of autonomy: the Kantian ethics of care' Nursing Philosophy 
3, 133-143  
 25
Parker R and Aggleton P 2002 HIV/AIDS-related stigma and discrimination: a 
conceptual framework and an agenda for action The Population 
Council/Horizons Program, Washington DC 
Proctor J D 1999 'Introduction: overlapping terrains' in Proctor J D and Smith D M 
(eds) Geography and ethics: journeys in a moral terrain Routledge, 
London 
Robinson F 1999 Globalizing care: ethics, feminist theory, and 
international relations Westview, Boulder CO 
Robson E 2001 'Interviews worth the tears? Exploring dilemmas of research with 
young carers in Zimbabwe' Ethics, Place and Environment 4(2), 135-142  
Ruiz P O and Vallejos R M 1999 'The role of compassion in moral education' Journal 
of Moral Education 28(1), 5-17  
Schoepf B G 2001 'International AIDS research in anthropology: taking a critical 
perspective on the crisis' Annual Review of Anthropology 30, 335-361  
Silk J 1998 'Caring at a distance' Ethics, Place and Environment 1(2), 165-182  
Silk J 2000 'Caring at a distance: (im)partiality, moral motivation and the ethics of 
representation' Ethics, Place and Environment 3(3), 303-309  
Silverstein K 1999 Millions for Viagra, pennies for diseases of the poor, The 
Nation, Nairobi 
Smith D M 1994 'On professional responsibility to distant others' Area 26(4), 359-367  
Smith D M 1997 'Back to the good life: towards an enlarged conception of social 
justice' Environment and Planning D: Society and Space 15, 19-35  
Smith D M 1998 'How far should we care? On the spatial scope of beneficence' 
Progress in Human Geography 22(1), 15-38  
Smith D M 1999 'Conclusion: towards a context-sensitive ethics' in Proctor J D and 
Smith D M (eds) Geography and ethics: journeys in a moral terrain 
Routledge, London 
Solomon R C 1996 'Emotions, ethics, and the 'internal ought'' Cognition and 
Emotion 10(5), 529-550  
Solomon R C 1997 'Beyond ontology: ideation, phenomenology and the cross 
cultural study of emotion' Journal for the Theory of Social Behaviour 
27(2/3), 289-303  
Taylor B 2001 'HIV, stigma and health: integration of theoretical concepts and the 
lived experiences of individuals' Journal of Advanced Nursing 35(5), 792-
798  
UNAIDS 2000 Report on the global HIV/AIDS epidemic. UNAIDS, Geneva 
 26
UNAIDS 2001 Advocacy for action on stigma and HIV/AIDS in Africa UNAIDS 
Intercountry Team for Eastern and Southern Africa, Pretoria 
UNAIDS 2004 Report on the global HIV/AIDS epidemic. UNAIDS, Geneva 
USAID/UNICEF/UNAIDS 2002 Children on the brink 2002: a joint report on 
orphan estimates and program strategies USAID/UNICEF/UNAIDS 
Valentine G 2003 'Geography and ethics: in pursuit of social justice - ethics and 
emotions in geographies of health and disability research' Progress in 
Human Geography 27(3), 275-380  
Weiser J 1999 'Adapting traditional healing practices' AIDS Action 46, 7  
Wenstop F and Magnus P 2001 'Value focused rationality in AIDS policy' Health 
Policy 57(1), 57-72  
Whatmore S 1997 'Dissecting the autonomous self: hybrid cartographies for a 
relational ethics' Environment and Planning D: Society and Space 15, 37-
53  
WHO Global Programme on AIDS 1991 'AIDS: credible messengers' World AIDS 
Day Features 1, 1-3  
Widdowfield R 2000 'The place of emotions in academic research.' Area 32(2), 199-
208  
Wiener J 2001 'Confidentiality and paradox: the location of ethical space' Journal of 
Analytical Psychology 46(3), 431-442  
Wilton R D 2000 ''Sometimes it's OK to be a spy': ethics and politics in geographies 
of disability' Ethics, Place and Environment 3(1), 91-97  
Young L and Ansell N 2003 'Fluid households, complex families: the impacts of 
children's migration as a response to HIV/AIDS in southern Africa' 
Professional Geographer 55(4), 464-479  
 
                                                 
Notes 
1 Indeed, moral universalism is arguably itself historically and culturally specific: Kant has 
been criticised as Eurocentric in his advocacy of the autonomous self (Paley 2002), and 
Habermas has identified how the universalist notion of ‘basic rights’ emerged in Europe over 
the past two to three centuries (Smith 1999). 
2 e.g. UK National Children’s Bureau Guidelines for Research 1993, Centre of the Child and 
Society, University of Glasgow Code of Practice for Research Involving Children (cited in 
Cree et al 2002), Priscilla Alderson’s (1995) Listening to Children.  
 
 
 
 
 
 27