EXPLOITATIONS AND THEIR COMPLICATIONS: THE NECESSITYOF IDENTIFYING THE MULTIPLE FORMS OF EXPLOITATION INPHARMACEUTICAL TRIALSbioe_1847 251..258

JEREMY SNYDER

Keywordsexploitation,pharmaceutical testing,research ethics

ABSTRACTHuman subject trials of pharmaceuticals in low and middle income coun-tries (LMICs) have been associated with the moral wrong of exploitation ontwo grounds. First, these trials may include a placebo control arm evenwhen proven treatments for a condition are in use in other (usuallywealthier) parts of the world. Second, the trial researchers or sponsors mayfail to make a successful treatment developed through the trial available toeither the trial participants or the host community following the trial.

Many commentators have argued that a single form of exploitation takesplace during human subject research in LMICs. These commentators donot, however, agree as to what kind of moral wrong exploitation is or whenexploitation is morally impermissible. In this paper, I have two primarygoals. First, I will argue for a taxonomy of exploitation that identifies threedistinct forms of exploitation. While each of these forms of exploitation hasits critics, I will argue that they can each be developed into plausibleaccounts of exploitation tied to different vulnerabilities and different forms ofwrongdoing. Second, I will argue that each of these forms of exploitationcan coexist in single situations, including human subject trials of pharma-ceuticals. This lesson is important, since different forms of exploitation in asingle relationship can influence, among other things, whether the relation-ship is morally permissible.

Human subject trials of pharmaceuticals in low andmiddle income countries (LMICs) have been associatedwith the moral wrong of exploitation on two grounds.First, these trials may include a placebo control arm evenwhen proven treatments for a condition are in use inother (usually wealthier) parts of the world. Second, thetrial researchers or sponsors may fail to make a successfultreatment, developed through the trial, available to eitherthe trial participants or the host community after thetrial. In both cases, researchers and their sponsors areaccused of wrongfully taking advantage of or exploitingthe members of these relatively poor and vulnerablecommunities.

Many commentators have argued that a single form ofexploitation takes place during human subject research inLMICs. These commentators do not, however, agree as

to what kind of moral wrong exploitation is or whenexploitation is morally impermissible.1 While accounts ofexploitation in human subject trials admit of many varia-tions, they can be divided into two broad categories: 1)Exploitation as unfairly taking advantage of others,where the gains produced through the interaction aredistributed unfairly; and 2) Exploitation as merely usingothers for one’s own ends.2

In this paper, I have two primary aims. First, I willargue for a taxonomy of exploitation that identifies twokinds of exploitation tied to different forms of unfairness

1 For a selection of these arguments, see J. Hawkins & E. Emanuel, eds.2008. Exploitation and Developing Countries: The Ethics of ClinicalResearch. Princeton, NJ: Princeton University Press.2 C. Grady. Vulnerability in Research: Individuals with Limited Finan-cial and/or Social Resources. J Law Med Ethics 2009; 37(1): 19–27.

Address for correspondence: Jeremy Snyder, Faculty of Health Science, Simon Fraser University, Blusson Hall 10516, 8888 University Drive, BurnabyBC, V5A 1S6 CANADA. Phone: 778.782.3258, Email: jeremycsnyder@sfu.ca

bs_bs_banner

Bioethics ISSN 0269-9702 (print); 1467-8519 (online) doi:10.1111/j.1467-8519.2010.01847.xVolume 26 Number 5 2012 pp 251–258

© 2010 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

and one, specific kind of exploitation tied to the mere useof others. Each of these forms of exploitation has itscritics and in the space of this paper I will be unable todevelop them fully. Nonetheless, I will make the case forthe plausibility of each account of exploitation and arguethat they are each tied to different vulnerabilities anddifferent forms of wrongdoing. This taxonomy is neces-sary given the tendency in discussions of exploitation inhuman subject trials not to specify the form of exploita-tion alleged or to assume the existence of a single form ofexploitation. Second, I will argue that these forms ofexploitation can coexist in single interactions, includinghuman subject trials. This lesson is important since dif-ferent forms of exploitation in a single relationship caninfluence, among other things, whether the relationship ismorally permissible.

CHARGES OF EXPLOITATION IN HUMANSUBJECT RESEARCH

In order to illustrate common uses of the charge ofexploitation in the context of human subject trials inLMICs, I will describe a case that appears frequently inthe literature on exploitation in human subject research.My aim is not to determine whether the actions describedin this case were exploitative but, rather, to help explorewhat kind of moral wrong is identified by the charges ofexploitation that have been used in response to this caseand others like it.

Many premature infants are born with insufficient sur-factant in their lungs, leading to respiratory distress syn-drome (RDS). Surfactant replacement therapy can beused to treat RDS. Four surfactants were approved bythe US Federal Drug Administration (FDA) between1990 and 2000. In 2000, the US pharmaceutical companyDiscovery Labs sought approval to conduct a phase IIItrial of a new surfactant called Surfaxin. Discovery Labsand the FDA agreed that a noninferiority trial comparingSurfaxin to existing treatments would not yield suffi-ciently high quality data. Instead, a placebo-controlledtrial (PCT) was proposed for 650 prematurely borninfants in Bolivia. All participating infants would receiveendotracheal tubes, ventilators, and antibiotics – treat-ments known to improve the chances of surviving RDS.Some of these treatments were at the time unavailable toinfants in the local community not enrolled in the trial.The hospitals selected for this multicenter trial werechosen because they were not able regularly to providetheir patients with surfactant therapy. Therefore, thoseinfants in the placebo arm of the trial would not receiveworse treatment than that which they would havereceived otherwise. In fact, insofar as they did not haveaccess to endotracheal tubes, ventilators, and antibiotics,their treatment would be better. While Discovery Labs

had some discussions about making Surfaxin available toparticipating hospitals at reduced cost if the therapy wasapproved, no such agreement was reached.3

One element of particular note in this case is that it wasarguably beneficial to and voluntary for all partiesinvolved in the trials. While human subject trial inLMICs pose difficulties in obtaining fully informedconsent, there is nothing about the case, as it is described,that demonstrates a failure to obtain consent. Moreover,even those participants who would have been enrolled inthe placebo arm of the trial may have received better careand better access to medical resources than they wouldotherwise have received. Even if these benefits were out-weighed by the risks of trial participation, all trial par-ticipants received, ex ante, a chance of enrollment in theactive arm of the trials, along with all of the potentialbenefits this opportunity created. There was a strongchance that this trial would therefore create a favorablebalance of costs and benefits for the trial participants.

Putting aside outright harms or a failure of consent, twoprimary pathways for exploitation have been alleged incases of this kind. First, the use of a placebo control arm,when proven treatments exist and are the standard of carein wealthier communities, has been said to be exploit-ative.4 The use of a PCT when proven treatments are avail-able in wealthier communities raises concerns of exploita-tion insofar as researchers take advantage of the povertyof research subjects in LMICs to enroll them in trials thatwould not be permitted in wealthier communities. Thislower standard of care works to the advantage of research-ers since it affords them the opportunity to conduct theirresearch more cheaply, quickly, and effectively than mightbe the case if a higher standard of care – and thus an activerather than placebo control – were used.5

Second, the accrual, or lack thereof, of benefits fromthe trial for trial participants and members of the hostcommunity has been cited as a source of potential exploi-tation. In the Surfaxin case, the expense of this drug, ifsuccessful, would have placed it far out of the reach of thecommunities in which it was being tested, given theirlimited per capita expenditures on healthcare. While Dis-covery Labs did initiate discussions to provide Surfaxinto the host communities at reduced cost if the drugproved successful, a firm agreement was not reached.6

Therefore, while the researchers and their sponsors wouldbenefit from the trial, the members of the communitiesparticipating in the trial would benefit very little by

3 Hawkins & Emanuel, op. cit. note 1.4 P. Lurie & S. Wolfe. 2007. The Developing World as the ‘Answer’ tothe Dreams of Pharmaceutical Companies: The Surfaxin Story. InEthical Issues in International Biomedical Research. J. Lavery, C. Grady,E. Wahl & E. Emanuel, eds. New York: Oxford University Press.5 M. Angell. Investigators’ Responsibilities for Human Subjects inDeveloping Countries. N Engl J Med 2000; 342(13): 967–969.6 Hawkins & Emanuel, op. cit. note 1.

252 Jeremy Snyder

© 2010 Blackwell Publishing Ltd.

comparison. For some commentators, the disproportion-ately great benefit received by the researchers serves asclear evidence that exploitation has occurred.7

The use of PCTs and failure to make any benefits fromthe trial available to trial participants and the host com-munity motivate much of the concern with exploitation inhuman subject trials in LMICs. However, the charge ofexploitation in response to these actions is frequentlypoorly articulated and the specific type of moral wrongthat is charged to have occurred is often unclear. This lackof clarity is worrisome. If it is not clear what form of moralwrong is meant by the charge of exploitation, then debatesover whether specific practices in human subject researchare exploitative will devolve into different parties talkingpast one another. Moreover, without a clear understand-ing of how so-called exploitative relationships in humansubject research are exploitative, it will be impossible torespond adequately to those commentators who claimthat these practices are not morally problematic.8 Merelyreiterating that they are exploitative will at best beg thequestion against these critics. Therefore, a closer exami-nation of the charge of exploitation in the context ofhuman subject research in LMICs is needed.

EXPLOITATION AS UNFAIRNESS

The wrongness of exploitation is most commonly tied tothe unfairness of the relationship, where unfairness isunderstood specifically in terms of the distribution of theresources created through an interaction.9 Exploitation,then, can be defined as taking unfair advantage ofanother person. In the context of human subject trials inLMICs, the worry is that the use of PCTs or the failure toprovide sufficient benefits for the trial participants andhost community will amount to an insufficient distribu-tion of the benefits of the trial among vulnerable partici-pants.10 If so, pharmaceutical researchers can be said tohave exploited these participants in the placebo arm ofthe trial.11 These commentators are often not clear,however, as to what kind of distribution would count asfair and why. Two relatively detailed accounts ofexploitation as fairness are available, which I will divideinto micro and macro fairness views.

Micro fairness

One strategy for interpreting the concept of fairness as itapplies to exploitation is explicitly to limit the range ofcontextual factors that will determine whether a distribu-tion of resources is unfair or not. On this view, externalfactors such as global economic structures, past injusticesagainst individuals, and the socioeconomic position ofvarious actors will not factor into the determination ofwhether the division of resources created by the interac-tion is unfair. Because this approach to understandingfairness seeks to limit the scope of the factors used in thisdetermination, I refer to it as the micro fairness approach.

In some cases, the exact standard for measuring fair-ness remains vague. Alan Wertheimer gives a very generalaccount of micro fairness, where ‘A exploits B when Agains more than A should (or B gains less than B should)from the transaction.’12 Other commentators couch thefairness standard in the language of inequity with theattitude that, whatever standard of fairness or inequityone endorses, the use of placebo controls and failure toguarantee benefits to the host community marks a clearviolation of the requirements of fairness.13 In other cases,commentators worry that benefits received by theresearchers and, especially, pharmaceutical companies,can be in some sense ‘disproportionate’ and thereforeexploitative.14

Other, more detailed accounts of micro exploitationdraw on Wertheimer’s earlier, groundbreaking work onthe concept of exploitation, where the unfairness found inmany cases of exploitation is tied to a deviation from thedistribution of benefits that would be achieved underhypothetical fair market conditions.15 This adaptation ofWertheimer’s approach has the advantage of clarity,where an explicit standard is given for measuring fairnesswithin an interaction. However, it suffers from severalshortcomings. First, it is not clear that human subjectresearch should be evaluated by the standards of fairnessused in the marketplace. While human subject researchdoes not aim to provide basic health care for researchsubjects, it does have an enormous impact on thosesubjects’ health and wellbeing, meaning market norms offairness may not be appropriate. Second, critics havecharged that a hypothetical fair market standard – andmore generally, micro fairness approaches that limit the

7 L.H. Glantz et al. Research in Developing Countries: Taking‘Benefit’ Seriously. Hastings Cent Rep 1998; 28.8 See, for example, R. Temple & S. Ellenberg. Placebo-Controlled

Trials and Active-Control Trials in the Evaluation of New Treatments.Part 1: Ethical and Scientific Issues. Ann Intern Med 2000; 133(6):455–463.9 A. Wertheimer. 1996. Exploitation. Princeton, NJ: Princeton Univer-

sity Press.10 S. Gbadegesin & D.Wendler. Protecting Communities in HealthResearch from Exploitation, Bioethics 2006; 20(5): 248–53.11 R. Crouch & J. Arras. AZT Trials and Tribulations. Hastings CentRep 1998; 28(6): 26–34.

12 A. Wertheimer. Exploitation in Clinical Research. In Hawkins &Emanuel, op. cit. note 1, p. 74.13 A. Ballantyne. HIV International Clinical Research: Exploitationand Risk. Bioethics 2005; 19(5–6): 476–491.14 D. Orentlicher. Universality and its Limits: When Research EthicsCan Reflect Local Circumstances, J Law Med Ethics 2002; 30(3): 403–10.15 E. Emanuel and the participants in the 2001 Conference on EthicalAspects of Research in the Developing Countries. Moral Standards forResearch in Developing Countries: From ‘Reasonable Availability’ to‘Fair Benefits’. Hastings Cent Rep 2004; 34(3): 17–27.

Exploitations and Their Complications 253

© 2010 Blackwell Publishing Ltd.

scope of fairness – fail to take into account forms ofunfairness created by the background institutional struc-tures within which these transactions take place.16

Macro fairness

The failure of micro fairness exploitation to account forbackground fairness motivates the use of a macro fairnessaccount of exploitation. A macro fairness account ofexploitation will include background, structural inequali-ties and injustices when determining whether the distri-bution of resources resulting from an interaction is fair.While he does not use the term ‘exploitation’, ThomasPogge argues that it is morally problematic to takeadvantage of social injustice, particularly if one has had arole in creating or perpetuating that injustice.17 He notesthat in the Surfaxin case, Discovery Labs could justly beaccused of perpetuating poor access to essential medi-cines in LMICs by being part of a network of pharma-ceutical companies that have encouraged a patentingsystem that reduces access to essential medicines in thesecountries. But even when the pharmaceutical company orresearchers do not directly perpetuate background injus-tice, this injustice can create a range of vulnerabilities thatcan unfairly be taken advantage of by placing persons ina worse bargaining position and in a situation of greaterdesperation than would otherwise have been the case.18

While many vulnerabilities can be taken advantage ofwithout fault – consider a surgeon benefiting from apatient’s blocked artery – the argument is that exploitersare not entitled to benefits accrued in virtue of exploiteesbeing disadvantaged by injustice. By way of comparison,a customer at a pawn shop may not be entitled to retainstolen goods, even if these goods were not stolen directlyby her and not known to be stolen at the time ofpurchase.19

Just as the micro fairness account can offer a hypo-thetical fair market price against which to measure fair-ness, a similar approach can be taken by proponents ofthe macro fairness account. A researcher or her sponsor,concerned about micro fairness exploiting research sub-jects, could choose to offer terms for a trial similar tothose that would take place in a hypothetical fair market.Even if bargaining asymmetries would allow theresearcher to bargain for more personally favorableterms of exchange, she can restrict her demands on poten-tial research subjects more in keeping with the terms that

would be agreed to under more equitable bargaining con-ditions. Similarly, a researcher or her sponsor might alsorestrict the demands on the victims of injustice to create ahypothetical just exchange price. This is essentiallyPogge’s strategy when he argues that Discovery Labs‘ought to treat Bolivian infants as if they were citizens ofa just country and therefore entitled to some basic healthcare (presumably including some existing RDS treat-ment) as a matter of course’.20 If the research subjects areimagined to have an entitlement to basic health care, thenat the very least the use of PCTs when proven treatmentsare available in wealthier parts of the world will count astaking unfair advantage of global injustice.

EXPLOITATION AS THE MERE USEOF OTHERS

While exploitation is often defined as taking unfairadvantage of others, the wrongness of exploitation neednot be tied to the unfairness of the distribution of thesocial surplus generated by an interaction. A broadlyKantian conception of exploitation can tie the wrongnessof exploitation to the mere use of others. The mere use ofothers often involves coercion or other deficiencies in anindividual’s consent. This is not the concern motivatingexploitation as mere use, however, as this form of exploi-tation can be mutually voluntary. Instead, the concern isthat relatively well-off researchers and sponsors fail todischarge a specified duty of beneficence to their researchsubjects.

In the case of health-related research, research subjectsmay enter into the relationship out of a desire to meetsome of their basic health needs. Researchers and theirsponsors may then be in a position to decide whethervoluntarily to cede some of the benefits derived from therelationship to better provide for the health needs of thesubjects and the host community than these subjectswould freely agree to, given their desperation for accessto any additional healthcare.21 By failing to provide phar-macologically active treatment in the control arm of thetrial or to secure access to the benefits of the trial for thetrial participants and host community, the researcher orsponsor may fail to discharge a duty of beneficence tohelp provide for the basic needs of others. These actionscan indicate that the researcher is acting on a maxim thatdisregards the basic needs of other rational humanagents.

Typically, the duty of beneficence is understood to beimperfect, giving the agent latitude as to when, how, andto whom it will be discharged, particularly if there are

16 R. Sample. 2003. Exploitation: What It Is and Why It’s Wrong.Lanham, MD: Rowman & Littlefield.17 T. Pogge. Testing Our Drugs on the Poor Abroad. In Hawkins &Emanuel, op. cit. note 1.18 D. Resnik. Exploitation in Biomedical Research. Theor Med 2003;24(3): 233–259.19 H. Steiner. A Liberal Theory of Exploitation. Ethics 1984; 94(2):225–241.

20 Pogge, op. cit. note 27, p. 113.21 N. Jecker. Exploiting Subjects in Placebo-Controlled Trials. Am JBioeth 2002; 2 (2): 19–20.

254 Jeremy Snyder

© 2010 Blackwell Publishing Ltd.

many people whose basic needs are not provided for.Understood in this way, the imperfect duty of beneficenceis sensitive to the context in which the researcher findsherself and does not demand that she give a fixed amountor even a fixed ratio of her resources to her researchsubjects. This flexibility is appealing since there is goodreason to think that the specific circumstances of eachrelationship – and especially the needs of the subjects andcapacity of the researcher to meet these needs – will behighly relevant to determining whether exploitation hastaken place in a specific case.22

However, there is good reason to think that an imper-fect duty of beneficence goes too far in granting latitudeto the researcher in determining how to discharge thisduty, particularly insofar as it is used to explain thewrongness of exploitation. The concern is that treatingexploitation as the violation of an imperfect duty fails toacknowledge the ways in which a relationship with par-ticular other persons can create new demands on ourresources. Exploitation is typically criticized as disregard-ing the needs of a particular person. An imperfect duty ofbeneficence allows the researcher to fully discharge herduty to attend to the needs of others by helping anyperson in need anywhere in the world. Therefore, animperfect duty of beneficence is not able to attend to theintuition that it is the use of another person for one’s ownbenefit that creates the potential for a special form ofwrongdoing.

Why, then, should we think that a researcher’s interac-tion with her research subjects creates special obligations?Jennifer Hawkins identifies a particular form of exploita-tion tied to what she calls ‘positive obligation flouting’where one person potentially harms another by taking anaction that leaves her below a moralized baseline of whatis owed to her.23 As opposed to a purely descriptive base-line of the wellbeing of each party before the interaction,a moralized baseline is set by positive obligations to leaveanother person with a set level of wellbeing. Therefore, aninteraction might be mutually beneficial when comparedto a descriptive baseline of the welfare of each party priorto the transaction, but leave one party worse off than sheshould be when compared to a moralized baseline ofwhat is owed to her from the interaction.

Hawkins stresses that researchers do not have the samepositive obligations to their research subjects as physi-cians have to their patients. The ‘therapeutic miscon-ception’ is common within the researcher-subjectrelationship and researchers must take care to clarifytheir role with regard to the health of their subjects.Nonetheless, positive obligations can be created by one’s

relationship with, or position relative to, those in need. Inparticular, a researcher might be best positioned to meetthe needs of her research subjects in virtue of her engage-ment with them through the research process.

Alternatively, the duty of beneficence can be shaped bythe roles and relationships that make up the context of aninteraction or what Alisa Carse and Margaret Little callthe ‘normative enterprise’ in which parties becomeengaged.24 Carse and Little argue that exploitation cantake place when an individual takes advantage of a vul-nerability in others that she is forbidden from takingadvantage of by virtue of the normative enterprise inwhich she is engaged. In the context of clinical trials,Carse and Little charge that the relationship betweenresearchers and their subjects should not be governed bythe norms of the marketplace, where each participantseeks to maximize her own utility when contracting withone another. Should the researcher take full advantage ofthese vulnerabilities in order to maximize her utility, shecan reasonably be accused of exploiting her researchsubjects.

While the views of Hawkins and Carse and Littlediffer, each gives arguments for conceiving of exploita-tion as a failure of a perfect form of the duty of benefi-cence. Moreover, these two views are compatible insofaras they are understood as outlining some of the factorsthat help to specify the duty of beneficence. Otheraccounts of exploitation as mere use build on this workby helping to explicate the factors that shape the duty ofbeneficence and its relationship to exploitation.25

TOWARD A COMPREHENSIVE ACCOUNTOF EXPLOITATION

I would now like to argue that each of the three viableaccounts of exploitation – micro fairness, macro fairness,and the perfect form of exploitation as mere use – isnecessary to understand fully and to avoid the danger ofexploitation in human subject trials in LMICs. Each ofthese accounts of exploitation is compatible with theothers and, taken together, they offer a more completepicture of the various means by which the vulnerabilitiesof potential research subjects in LMICs can wrongfullybe taken advantage of by researchers and their sponsors.

While most accounts of exploitation claim to providea single, unified account of the moral wrong ofexploitation, each of the three forms of exploitation thatI have identified is associated with different kinds of vul-nerabilities that may be wrongfully taken advantage of.

22 See A. Siegel. Kantian Ethics, Exploitation, and Multinational Clini-cal Trials. In Hawkins & Emanuel, op. cit. note 1.23 J. Hawkins. Exploitation and Placebo Controls. In Hawkins &Emanuel, op. cit. note 1.

24 A. Carse & M. Little. Exploitation and the Enterprise of MedicalResearch. In Hawkins & Emanuel, op. cit. note 1.25 See Sample, op. cit. note 16; and J Snyder. Needs Exploitation.Ethical Theory Moral Pract 2008; 11(4): 389–405.

Exploitations and Their Complications 255

© 2010 Blackwell Publishing Ltd.

Taken together, these accounts of exploitation build onthe moral intuition that it can be wrong to press one’sadvantage or to take full advantage of others’ vulnerabili-ties in certain settings. 1) Micro fairness exploitationestablishes that localized asymmetries in the bargainingposition of actors can create a vulnerability that, if fullytaken advantage of, allows the better positioned party tosecure much more of the social surplus created by theinteraction than would be the case in a more competitive,better balanced, or more equal relationship. 2) Macrofairness exploitation describes how background institu-tions, particularly when these institutions are unjust,create long-term asymmetries in the bargaining power ofactors, tied to their class, race, geographic position,gender, and other factors. When more privileged actorstake advantage of these vulnerabilities, they can secure anunfair portion of the social surplus created through aninteraction. 3) Exploitation as mere use shows that wecan fail to meet our specified obligation of beneficencetoward others, instead retaining resources that are owedto others for our own use. Contextual factors such asone’s role, proximity to others, resources available tooneself, and the normative enterprise in which one isengaged can all create a context in which the vulnerabili-ties of specific other persons should be reduced ratherthan merely taken advantage of.

Each of these kinds of vulnerability can exist singly ortogether, meaning that these forms of exploitation cancoexist in individual contexts. Both micro and macrofairness exploitation identify factors that allow individu-als to secure an unfair portion of the social surplus of aninteraction. While proponents of each account may takethe view that their preferred version identifies the realwrong of exploitation, each of these forms of exploitationcan be understood as identifying distinct kinds ofvulnerabilities that generate exploitable bargainingasymmetries.

The complaint of some micro fairness accounts againstthe macro fairness accounts is that they demand toomuch of individuals by requiring that the potentialexploiters make up for the disadvantages created by insti-tutional injustice. However, macro fairness accounts canoffer less demanding duties. For example, instead ofrequiring that one refrain completely from benefitingfrom unjust institutions, an account of macro fairness canmaintain that one’s connection to unjust institutionshelps to determine and limit the duty to address injus-tice.26 While micro fairness might demand that all local-ized unfairness be neutralized, macro fairness can merelyentail a positive obligation by actors to mitigate theeffects of injustice and to reform unjust institutions. Thevulnerabilities behind micro and macro fairness exploita-

tion, then, can each demand very different responsesgiven their different underlying vulnerabilities. These dif-ferences help to make the case that each kind of vulner-ability should be associated with a distinct kind of moralwrong and distinct kind of exploitation.

I have argued that a violation of the duty to addressstructural injustice, specified by one’s connection to theseinstitutions, can explain the wrongness of macro fairnessexploitation. This duty can be coupled with an obligationof researchers, grounded in a perfect duty of beneficence,to address the basic needs of their research subjects – theviolation of which can give rise to exploitation as mereuse. The role of the researcher, then, can generate specificduties to address specific vulnerabilities, whether they arethe bargaining asymmetries generated by unjust institu-tions or the shortfalls in the basic needs of others. Whilethese two vulnerabilities will often be found together,they are separable. As with micro fairness, macro fairnessfocuses on the distribution of resources, meaning thatunfairness can be decoupled from the basic needs ofothers. Exploitation as mere use, however, focuses solelyon the inability of others to meet their basic needs,whether these shortcomings are caused by structuralinjustice or have other causes.

There will be interactions in which each form of exploi-tation is present. If we return to the Surfaxin case, there isgood reason to think that the design of this trial createdthe potential, as least, for each form of exploitation. Thepotential for exploitation as mere use and macro fairnessexploitation appear the most straightforward. It is certainthat the basic health needs of the infants in the study werenot being met, given their complete lack of access to anysurfactants and the low per capita expenditures on healthcare in Bolivia. Insofar as the proximity, roles, and nor-mative enterprise in which the researchers were engagedhelped to specify each researcher’s duty of beneficence, afailure to provide benefits to the host community and theuse of a PCT could both violate this duty.

A swath of global institutions have been claimed to beresponsible, in part, for the poor health outcomes of thetrial participants. These could include very general insti-tutions like colonialism, a history of neoliberal economicpolicies, and anti-democratic political interventions inSouth America.27 Insofar as this is the case, individualswill have a political responsibility to reform these institu-tions, in keeping with their connection to them. More tothe heart of this discussion, pharmaceutical companiescan rightly be held liable for promoting internationalintellectual property agreements such as the agreementon Trade Related Aspects of Intellectual Property Rights(TRIPS) that reduce access to essential medicines in

26 See I. Young. Responsibility and Global Justice: A Social Connec-tion Model. Soc Philos Policy 2006; 23(1): 102–130.

27 T. Pogge. 2002. World Poverty and Human Rights. Malden, MA:Polity.

256 Jeremy Snyder

© 2010 Blackwell Publishing Ltd.

countries like Bolivia.28 To the degree that these institu-tions hamper the bargaining position of the parents in theSurfaxin case, researchers can be said to exploit thesevulnerabilities, both through the failure to grant fairerterms of access to the benefits of the trial to the hostcommunity and refusal to use a pharmacologically activecontrol arm for the trial.

Finally, researchers may exploit trial participants bythe standards of micro fairness as well, depending on theaccount of fairness being used. While I maintain that it isdeeply problematic to apply a hypothetical fair marketstandard to this case, one could take the use of a PCT andfailure to make Surfaxin widely available to the hostcommunity as strong evidence that the researchers andtheir sponsors received too great an advantage from theinteraction due to an asymmetrical bargaining relation-ship between a small cadre of researchers and the many,deeply impoverished communities that are desperate foreven the limited benefits that could be accrued throughhosting a pharmaceutical trial.

Having shown that the presence of each of the threeforms of exploitation is plausible in the Surfaxin case, Iwould now like to argue that the potential for multipleforms of exploitation in a single context is crucial todetermining when exploitation does, in fact, take place.Each of these forms of exploitation can influence oneanother and, partially, determine when and whether eachtakes place. The influence of the three forms of exploita-tion on one another is important because the most plau-sible versions of each of the three accounts of exploitationmaintain that either: 1) Exploitation can be morally per-missible, all things considered, given competing ethicalconsiderations; or 2) Whether a situation counts asexploitative depends on other contextual factors includ-ing other ethical obligations. Many proponents of thesethree forms of exploitation argue that either it is an openquestion of whether a mutually beneficial and voluntarybut exploitative relationship should be forbidden orwhether competing moral demands can outweigh theobligations of beneficence and fairness such that a seem-ingly exploitative relationship is not exploitative. In eachcase, the fact that mutually beneficial and voluntaryhuman subject trials do something to help the potentiallyexploited party speaks to the need to consider the conse-quences of setting a bar on these interactions too high.

Among those that see exploitation as permissible, allthings considered, in certain situations, the consequencesof the interaction are relevant to making this determina-tion. Wertheimer, for example, writes regarding the Sur-faxin case that ‘in the absence of good reasons to thecontrary, we should generally permit mutually advanta-geous and consensual exploitation to go forward,

although it may turn out that there are good reasons tointervene in the present case.’29 While this considerationof the consequences of the interaction need not take theform of maximizing utility, it does acknowledge that themoral wrong of exploitation is not the only moral con-sideration relevant to assessing what one ought to do in agiven case.

Alternatively, the determination of whether one hasengaged in exploitation can depend on the resourcesavailable to the potential exploiter and the consequencesof her actions. Carse and Little are clear that the use ofPCTs and the failure to make the treatment reasonablyavailable to the host community will typically be morallyimpermissible. However, they too grant that the use ofplacebos when proven treatments exist can be justifiedwhen ‘doing so is needed to help those in similar healthcircumstances (and not, say, because using this popula-tion provides an especially cheap way to collect cleandata for those who do not suffer the “useful” depriva-tion.’30 The full context in which the researcher findsherself, then, can include both the resources available toher and the consequences of her actions. This contexthelps determine whether exploitation takes place bydetermining, in the case of PCTs, whether the proposedresearch trial is the best available means of meeting thebasic needs of stakeholders and whether doing so wouldbe too demanding.

Under either of these views, the determination ofwhether an interaction is exploitative and morally imper-missible will entail a weighing of the moral wrong ofexploitation and other moral considerations. Crucially,this weighing process cannot be fully accomplished if wedo not recognize that multiple forms of exploitation mayoccur in single contexts and, therefore, multiple, distinctforms of wrongdoing need to be part of this weighingprocess. If, for example, the moral wrong of micro fair-ness exploitation is weighed against the positive conse-quences of the Surfaxin trial without considering themoral wrongs associated with macro fairness exploitationand exploitation as mere use, then we cannot be confidentthat we will arrive at a full consideration of the moralterrain of the case.

The lesson that I would like to emphasize by this dis-cussion is that the presence of multiple forms of exploi-tation will complicate the calculation as to whether aspecific form of exploitation occurs and/or is permissible.Even when developing full accounts of these variousforms of exploitation, we must be cognizant of the otherforms of exploitation insofar as each is prone to be rel-evant to the application of the others. Given that theconcept of exploitation seeks to explain when and whytaking advantage of the vulnerabilities of others is

28 T. Pogge. Access to Medicines. Public Health Ethics 2008; 1(2):73–82.

29 Wertheimer, op. cit. note 12, p. 100.30 Carse and Little, op. cit. note 24, p. 234.

Exploitations and Their Complications 257

© 2010 Blackwell Publishing Ltd.

morally problematic, it should not be surprising both thatdifferent vulnerabilities will give rise to different kinds ofwrongdoing and that these different vulnerabilities willoften reside within the same relationship. We make amistake if we forget this lesson and attempt to discussexploitation as if it denotes a single kind of wrongdoingor can be discussed without reference to the multipleforms that it can take. If these distinctions are not madeclear, discussions of exploitation in human subject

research are more likely to muddle than clarify theforms of wrongdoing that can take place in these rela-tionships.

Jeremy Snyder is an Assistant Professor in the Faculty of Health Sci-ences at Simon Fraser University in Burnaby, British Columbia. Hisresearch interests encompass ethical issues related to public and popu-lation health, including medical tourism, health worker migration, lowwage labour, ethical issues in human subject trials, and the concept ofexploitation.

258 Jeremy Snyder

© 2010 Blackwell Publishing Ltd.