JAMES H.

FLORY AND Global Health and the

PHILIP KITCHER
Scientific Research Agenda

I
In , million people died. Violence killed fewer than one million of
them; famine contributed to about six million deaths; more than
million died of some form of disease.1 Many of these illnesses were the
result primarily of old age and may have been unpreventable. That is
unlikely to be true, however, of over a million deaths from malaria, nearly
two million deaths from tuberculosis, nearly three million deaths from
AIDS, and over four million deaths from respiratory disease.2
To suggest that many of these deaths could have been prevented
might be to make one of two distinct claims. First, in the current state of
our knowledge, available techniques exist for responding to the diseases
mentioned above and for saving those who suffered from them. Second,
these are scourges that we might hope to overcome in the course of
future biomedical research. Part of the worlds burden of disease could
have been alleviated if those who died had had access to drugs or treat-
ments routinely available to others in different places. Another part
could have been lightened if there had been more thoroughgoing efforts
to discover methods of combating disease, methods that the actual
course of biomedical research has so far not yet found.

We are extremely grateful to two anonymous referees whose helpful comments on

earlier versions have enabled us to make substantial improvements; we also wish to thank
the editors for their valuable suggestions. We also note that the views expressed here are
the authors own and do not reflect the position of the National Institutes of Health or of
the Department of Health and Human Services.
. C.J.L. Murray and A. Lopez Evidence-Based Health Policy Lessons from the Global
Burden of Disease Study, Science (): .
. For data see the Full Report: World Health Report (Geneva: World Health
Organization, -), not only for but for other recent years as well.

by Blackwell Publishing, Inc. Philosophy & Public Affairs , no.

37 Global Health and the Scientific
Research Agenda

Our aim is to understand the relation between the ways that people
die and the kinds of weapons against disease that medicine has acquired
and that it seeks to extend. We shall try to give substance to the familiar
view that the diseases that contribute to the global burden of death are
not well-aligned with the dominant directions in biomedical research.
This misalignment, we claim, underwrites an obligation for individual
researchers to reorient their inquiries and for the institutions that
support research to alter their priorities. We begin with some facts about
the worldwide distribution of disease and the strategies for treating
those afflicted.

II
In the poorest nations, many people die prematurely from the diseases
we have mentioned. Before we think about possibilities for new tech-
nologies, it is worth asking if we could make better use of the knowledge
already available. Many nations use existing biomedical techniques to
provide effective control of tuberculosis and similar diseases. But most
of the countries where malaria, tuberculosis, respiratory infections, diar-
rhea, parasitic infestations, and so forth are rampant owe their trouble
not to the lack of twenty-first century medical technology but to the
absence of late nineteenth century sanitation. For example, one of the
most historically successful strategies for malaria control is to eliminate
stagnant water sources where mosquitoes breed. Even with emerging
malarial resistance to drugs, it is certain that the burden of the disease
could be greatly reduced by drainage projects combined with a more
pervasive and effective system of public health in the affected countries.
Tuberculosis is a similar case. Inconsistent use of antibiotics has
contributed to the spread of drug-resistant tuberculosis, but even most
resistant strains can be beaten by a public health system that applies
existing therapies quickly and thoroughly. As matters stand, millions of
tuberculosis (TB) patients receive incomplete courses of drugs, bringing
temporary respite at the cost of giving the TB bacillus an opportunity to
evolve antibiotic resistance.3 If well-funded, trained clinicians reached

. The classic example is Russia after the collapse of the Soviet system, where
overcrowded prisons provided the tuberculosis bacillus with extraordinary new ecological
possibilities. See Laurie Garrett, Betrayal of Trust (New York: Oxford University Press, ),
p. .
38 Philosophy & Public Affairs

all these patients, the emergence of resistance could be slowed and the
vast majority of tuberculosis deaths prevented without a single new
technological idea.
It is immediately obvious, however, that making existing technologies
more broadly accessible is far from simple. In malaria, Africa has a pro-
found source of suffering that could be greatly relieved with clinical care
that employs only procedures that are routine elsewhere (the courses of
antibiotics that are administered in affluent countries); in tuberculosis,
the Russian federation faces the same situation. But the infrastructure
needed to supply that care no longer exists in the regions in question;
neither does the funding nor political will to build it.4 For outsiders to
replace this infrastructurebuilding clinics, training physicians and
nurses, supplying policy, and doing all of this for an indefinite period of
timeamounts to taking responsibility for the public health of the
nations involved. Such intervention may be an enormously valuable
undertaking, but it is also very expensive and politically delicate. Devel-
oping new ways to fight tuberculosis is a difficult scientific problem, but
rebuilding the former USSRs tuberculosis control programs may be an
even harder political one.
It would thus be an error to think that, simply because a strategy for
combating a disease is available in a particular, privileged, region of the
world, medical research on alternative strategies is pointless. When the
task of exporting the technology to other contexts is beset with large
socio-political obstacles, the best means of bringing relief may be to find
treatment better adapted to those contexts. Thus, when we consider
biomedical research, well want to consider two sorts of goals: most obvi-
ously, ways of addressing diseases for which no treatment is available
anywhere, but also new methods for fighting diseases for which the only
available therapies cannot be exported to the contexts in which many of
the afflicted find themselves. In other words, one cant simply declare that

. Garretts discussion of an Ebola virus outbreak in Kwikwit, Zaire (Betrayal of Trust,

ch. ) is especially pertinent here. She describes vividly how the public health network in
Zaire was unable to provide the basic resources western medicine takes for grantedclean
water and electricity were both lacking at the Kwikwit hospital. Thanks to the efforts of a
group of doctors, including Zairois, representatives from the Center for Disease Control
and members of Mdicins sans Frontires, what might have been a devastating epidemic
was stopped. A few years later, however, the hospital was in the same appalling state in
which the MSF-CDC team had found it.
39 Global Health and the Scientific
Research Agenda

a disease is sufficiently researched just because it no longer has an impact

on the affluent world.
Lets now turn to the relationship between the ways that biomedical
research is currently directed and the global burden of disease. Best esti-
mates are that the total spent on health research of any kind is around
$ billion.5 The National Institutes of Health (NIH) alone will spend
almost all of a $ billion budget on medical research next year. The
pharmaceutical industry, overwhelmingly based either in the United
States or the European Union, will spend about $ billion. In all, less
than $ billion of funding originates from the poorer parts of the world
Africa, the Russian Federation, the rest of Asia, and South and Latin
America.6
One rather obvious way to measure the alignment between disease
burden and the directions of biomedical research is to compare the
amount of money invested in a disease with the proportion of the total
budget ($ billion) that a disease would receive if the allotment were
made strictly according to the diseases fraction of the total number of
deaths due to disease (approximately million). Well define a diseases
fair share as the product of the number of deaths attributable to the
disease and $ billion, divided by million.7 Malaria kills one million
people a yearincluding a high proportion of childrenso its fair share
is $. billion; in , the actual figure spent on malarial research was
about $ million.8 Tuberculosis is responsible for two million deaths
each year, and so its fair share would be $. billion; high estimates are
that it received $ million in .9 (So if malaria sufferers historically
receive only a twentieth of the scientific attention their plight merits,

. Sheila Davies, / Report on Health Research (Geneva: Global Forum for

Health Research, ).
. Monitoring Financial Flows for Health Research (Geneva: Global Forum for Health
Research, October ).
. This notion plainly construes disease burden in terms of deaths, and so ignores (or,
at least, underrepresents) those diseases that give rise to long periods of pain and disabil-
ity; we shall come to terms with this point below.
. J. Anderson, M. MacLean, and C. Davies, Malaria Research: An Audit of Interna-
tional Activity, Wellcome Trust Unit for Policy Research in Science and Medicine, .
. C. Michaud, C.J.L. Murray, and B. R. Bloom, Burden of DiseaseImplications for
Future Research, Journal of the American Medical Association, (): . Funding
has surely increased during the s, especially with the knowledge that strains of TB
resistant to the major antibiotics were surfacing in New York City. But even if funding has
tripled or quadrupled, it is still miniscule compared to the scope of the problem.
40 Philosophy & Public Affairs

victims of tuberculosis may only get a hundredth.) The single largest

infectious cause of death is respiratory infection, killing four million
people per year; the third largest cause is the general category of diar-
rhoeal diseases, which take . million lives a year. Their fair shares
would be nearly $ billion and $ billion, respectively; in fact, the esti-
mates award them about $ million.10
Little imagination is needed to come up with an explanation for the
disparities we have noted. Research dollars come almost entirely from
the wealthy parts of the world, and the suffering from malaria, tubercu-
losis, and a large number of other infectious agents happens elsewhere.
Nine hundred thousand of the . million who die from malaria in any
given year are in sub-Saharan Africa; malarial death rates in the United
States are negligible. In , tuberculosis killed between . and .
million people worldwide, with the best estimate falling near .
million.11 Over one million of these deaths occurred in six developing
countries: India, China, Indonesia, Pakistan, Bangladesh, and Nigeria;
by contrast, in the United States tuberculosis annually kills about
people. In all of Europe, tuberculosis takes an annual toll of about
,, which is a tiny fraction of the total even before one considers
that most of these deaths occur in the poorer nations of Eastern Europe,
particularly in the Russian Federation. Wealthier nations typically have
death rates measured in the hundreds.12
Tuberculosis and malaria are only the two clearest examples of a
problem that pervades disease research. The most succinct way of
describing this is the / gap: percent of humanitys burden of
disease receives only percent of the worlds health research
resources.13 The skewed distribution favors the diseases of the affluent
worldcancer, heart disease, diabetes. How much of the resources

. Ibid. These causes of death may not rate the most intense research attention either
because they already have good low-tech solutions (e.g., oral rehydration therapy works
for some diarrheas) or because they cluster many different diseases, each taking a rela-
tively small toll. Thus the marginal rate of return for them may not be as high as it is for a
single major killer, such as malaria or tuberculosis. Interestingly, the second objection
applies to cancer research.
. Full Report: World Health Report .
. Christopher Dye et al., Global Burden of Tuberculosis, Journal of the American
Medical Association (): .
. Report of the Ad Hoc Committee on Health Research Relating to Future Inter-
vention Options (Geneva: World Health Organization).
41 Global Health and the Scientific
Research Agenda

aimed at these diseases could be diverted without appreciable loss is a

serious empirical question; we suspect (but dont claim to know) that
some diversion would be possible without significantly lessening the
chances of success with respect to these diseases. The cost of the current
bias is that vaccines and new antibiotics that would make a tremendous
difference to quality of life in the nonaffluent world continue to go unde-
veloped, and we are confident that there will be a modification of the
existing distribution in which the expected gains from research into new
vaccines and antibiotics outweigh any expected losses from the support
withdrawn from cancer, heart disease, and so forth.14
Weve introduced the concept of a diseases fair share of research
resources, noted the discrepancy between the fair share (as we define it)
and the actual amount assigned to those diseases that mainly afflict
people in the poorest parts of the world, and have offered an obvious
explanation. But there are two obvious objections to the analysis pre-
sented so far, both of which would charge our talk of fair share with
persuasive definition. The first would emphasize that the direction of
scientific research cannot simply be a matter of the urgency of the prac-
tical problems; the second would question our measure of the global
disease burden. We consider each criticism in turn.
There are sometimes good reasons for giving a disease more than its
proportional share of research funds on the grounds that the available
avenues for investigating it are more plentiful or more easily traversed
than routes for tackling other diseases. Yet if a disease is both particu-
larly deadly and appears vulnerable to research into new forms of treat-
ment or prevention, it would seem wrong to shortchange it. Well now
argue that both malaria and tuberculosis are promising targets of

. If our conjecture here is incorrect, then the right strategy might be to increase the
total amount of support given to biomedical research to allow both for the retention of the
advantages of present approaches to affluent-world diseases, and for the additional ben-
efits of attention to the diseases that afflict poor nations. We would also note that some
fundingand its hard to estimate just how muchgoes to develop lifestyle drugs, sub-
stances meant to alter the users body in ways not intended to cure disease. Attempts to
overcome impotence, halt hair loss, reduce obesity, and remove wrinkles attract hundreds
of millions of research dollars each year. These problems are trivial in comparison with the
suffering wreaked by the deadly diseases of the poor world. (See Ken Silverstein, Millions
for Viagra, Pennies for the Poor, The Nation, July , , pp. .) Since our primary
concern in this article will be with public funding of science, rather than with biophar-
maceutical research, we shall not press this point.
42 Philosophy & Public Affairs

research, so that the proportionality approach that underlies our

concept of fair share cannot be rejected on the suggested basis.
The malarial parasite implicated in most cases (Plasmodium falci-
parans) is a complex organism with a complex life cycle that could be
interrupted in many ways: it is potentially vulnerable to antibiotics, to
control of its animal vector, and to vaccines. Tuberculosis is also a
promising subject for research, both because of the chance of finding
new antibiotics and because of possibilities for vaccines. As weve
already noted, an important role for technology is to provide ways for
fighting disease, even in contexts in which the public health infrastruc-
ture has broken down. New drugs that circumvent resistance can make
an important reduction in the short term, although it will be crucial to
ensure that the application of those drugs now doesnt sacrifice chances
for a more efficacious intervention laterit would be tragic if, because
of the public health deficiencies, new antibiotics were quickly to gener-
ate new resistant strains. Our greatest hope is that many under-
researched diseases, including malaria and tuberculosis, might one day
be controlled with vaccines, and we shall focus on the prospects for
vaccine development.
An ideal vaccineone that is effective in a single dosecan be used
to control, or even eradicate, a disease everywhere in the world. The
smallpox eradication campaign carried out in the s attests suffi-
ciently to the enormous potential of vaccines: for a cost of less than a
billion of todays dollars, an eleven-year program completely wiped out
a disease that was highly infectious and killed two million people each
year. Had smallpox eradication never taken place, that disease would
have killed tens of millions of people and would now be doing damage
comparable to that done by AIDS, malaria, or tuberculosis. Instead, it is
simply gone, even in parts of the world where other forms of public
health are in tatters.15
The molecular revolution in biology and biomedicine offers the
potential for leaps forward in both drug and vaccine design. Its most
visible symbolthe Human Genome Projectis often seen by
researchers as a prelude to advances in basic biology (increased abilities
to map and sequence genes and to engage in comparative analysis of
the functions of genes in different organisms are likely to enhance our

. Laurie Garrett, The Coming Plague (New York: Penguin, ), pp. .

43 Global Health and the Scientific
Research Agenda

understanding of intracellular metabolism and organismal develop-

ment); the public, and some outspoken medical researchers, envisage
the possibility of new techniques for addressing the major diseases
of the affluent worldcancer, of course, hypertension, diabetes,
Alzheimers, bipolar depression, and so forth. Yet the ability to sequence
rapidly, originally demonstrated on bacterial genomes, also enables us
to analyze the tricks of major pathogens. In principle, one can map and
sequence the genome, hunt for proteins that are exposed on the surface
of the pathogen, design a harmless vehicle that will stimulate the
production of antibodies that would neutralize the pathogen, and so
prepare the immune system against its attack. There could easily be
a huge (and not desperately expensive) project to analyze pathogen
genomes and to generate potential vaccines (which could, at least some-
times, be tested quite rigorously on animal models).
High technology efforts wouldnt always work, of course. The example
of AIDSone of the few infectious diseases threatening enough to the
affluent world to attract billions of research dollarshas been the target
of a very intensive research effort. Not only the price and difficulty of
distributing drugs, along with fears of HIVs developing drug resistance,
lead many researchers to believe that the only long-term solution to
AIDS will be a vaccine. The fact that our detailed knowledge of the HIV
genome hasnt yet translated into success in making a vaccine demon-
strates that theres no guarantee that sequencing pathogens will deliver
the goods. But one might also conclude, from the half-century that has
elapsed since the discovery of the molecular basis of sickle-cell anemia,
that molecular techniques are hopeless in the battle against genetic
disease. To use an analogy one of us has deployed elsewhere, we should
think of the pathogen sequencing project as giving us a vast number of
tickets in a number of lotteries; we cant predict where well be success-
ful, and how big the prizes well win, but wed be profoundly unlucky to
come up completely empty.16
In the case of malaria, several vaccine candidates are at or near Phase
I human trials, but its estimated that a useful malaria vaccine is at least

. See Philip Kitcher, The Lives to Come: The Genetic Revolution and Human Possibil-
ities (New York: Simon & Schuster, ). Our view is that the chances are actually signifi-
cantly higher from the pathogen sequencing/vaccine development project than from the
standard ways of applying information from the Human Genome Project.
44 Philosophy & Public Affairs

ten years away.17 This is probably optimistic. Without more funding, its
likely that promising leads wont be explored, so that the ten years will
balloon into twenty or more. In the case of tuberculosis, vaccine research
is hardly off the ground. Because of the high expense of treating the
disease, and its propensity to become drug resistant, a tuberculosis
vaccine is particularly desirable. But tuberculosis vaccine research
receives even less funding than malaria vaccine research, and vaccine
development has lagged accordingly. No new tuberculosis vaccines are
in clinical trials today.18 Thanks largely to the recent sequencing of the
TB bacillus genome, promising new avenues for research are open, but
the research is only in the earliest stages, and the funding needed to
carry it out with a serious chance of success isnt yet available. Indeed,
at current levels of funding, a vaccine may never be developed.
Weve tried to defuse one line of objection, one that stresses the need
to take research promise into account in the apportionment of funding.
We now turn to the second, which charges that weve operated with too
crude a measure of suffering by concentrating on numbers of deaths. It
would be correct to note the possibility that some diseases inflict more
suffering for a given mortality rate than do others: a disease that kills
some number of people late in life may cause much less suffering than
one that kills the same number of people but strikes earlier and inflicts
long periods of disability in many nonfatal cases. Extensive work has
been done, largely by the World Health Organization, so that these con-
siderations can figure in a more nuanced calculation of the burden of
disease.
Instead of simply assuming that mortality rates correspond directly to
the amount of suffering, the most common practice has been to intro-
duce a unit known as the DALY (Disability-Adjusted-Life-Year). The fun-
damental idea is that we should look for the number of years of life lost
because of the disease. Life is lost not simply through early mortality but
also through the discounted value of years of life lived with disability,
where the discounting is based on widely shared views about the value
one would attribute to a future year of ones life in the diseased state as

. Quest for Malaria Vaccine Revs Up, But Much Work Remains, Bulletin of the WHO
(): .
. Theres an old vaccine, BCG, that has been in use for over seventy years. It can
provide children with some protection against some strains. For the most prevalent form
of tuberculosis, adult pulmonary tuberculosis, its quite useless.
45 Global Health and the Scientific
Research Agenda

opposed to the value one would assign to a disease-free future year. As

the idea is developed, there are two main assumptions, one egalitarian
and the other age-inegalitarian. The egalitarian assumption takes a year
of human life to be of equal value regardless of location, race, social
background, or level of education, and also incorporates the claim that
death is premature if it occurs before a standard of . years for women
and years for men. The age-inegalitarian assumption does place a
different value on years of life at different stages of a human beings
existence: initially, the value of a year rises rapidly with age, peaking in
the mid-twenties, and thereafter gradually declines to about a third of
its maximum value by the age of . This assumption is explicitly based
on studies of how people actually value different stages of human life;
part of its rationale is the idea that adults of the most valued ages can
make the greatest contributions to their societies and to the well-being
of their dependents.19
We take the egalitarian assumption to be fully defensible, and, in par-
ticular, we regard objections to a uniform standard of premature death
as ungrounded. An obvious worry is that the egalitarian assumption sup-
poses that diseases that strike well after the end of the usual lifespan of
a human being in the non-affluent world exact a large number of DALYs.
Were we to revise the DALY measurement to incorporate the expected
lifespan in particular regions, however, then wed already be acquiescing
in a particular burden of disease as the norm for the inhabitants of those
regions, writing off a portion of their suffering as something that didnt
count. Affluent people who view it as a hardship to be denied health,
even in their seventies, have no reason to consider a loss of lifelife
unencumbered by disability, recallunimportant in someone else. If
people in poorer parts of the world cant expect long and active lives, that
doesnt make it less bad for them not to have them; indeed, as weve sug-
gested, that expectation itself represents part of the burden of disease
that afflicts them.20

. C. Michaud, The Global Burden of Diseaes and Injuries in , International

Social Science Journal (): .
. We suspect that those inclined to make the objection weve just briefly discussed
would have argued, in an earlier day, that the deaths of slum-dwellers or coal miners in
their own societies couldnt be considered premature because, after all, they lived as long
as they had any right to expect. We hope that noting the parallel will remove the urge to
make either complaint.
46 Philosophy & Public Affairs

Were far less confident about the age-inegalitarian assumption. For

present purposes, however, we dont need to resolve the issues that it
raises. Unlike the commitment to a uniform standard of premature
death, the age-inegalitarian assumption doesnt increase the DALYs
associated with diseases that afflict poor nations.
Our earlier conclusion was that, when human suffering is measured
by raw mortality rates, the distribution of biomedical research is skewed
with respect to the distribution of suffering. How do things stand when
we replace mortality rates with DALYs lost? Roughly the same. Tubercu-
losis and malaria continue to exact the majority of their cost in the
non-affluent world, and the harm they do relative to other causes is
approximately the same as on our earlier (mortality-based) estimate.
Tuberculosis causes million out of million deathsslightly less than
percent of deaths; it costs about . percent of DALYs, a small reduc-
tion but by no means enough to eradicate the spectacular inadequacy
of tuberculosis research funding. Malaria maintains the same signifi-
cance as before, moving from causing about percent of deaths to being
responsible for . percent of DALY loss. In these cases the two most
obvious ways of estimating disease burden yield the same conclusion:
research investments and global disease burden are mismatched, and
matters are no different when we turn to respiratory disease or diarrhea.
Interestingly, a DALY analysis does show new diseases of poor regions
that attract very little research attentionmost prominently those
parasitic infestations that do not kill but produce blindness or severely
impaired mobility. A forceful example is onchocerciasis, also known as
river blindness. This African infectious disease slowly blinds its victims
without killing them; the WHOs statistical analysis of its effects there-
fore shows a negligible number of deaths but an annual loss of over one
million DALYs.21
Its worth noting that weve focused only on the immediate costs of
disease in the poor parts of the world. In our judgment, the consequences
of tuberculosis, malaria, respiratory infections, diarrhea, and their ilk are
so economically expensive that the world would be richer and more
politically stable if we could eradicate them. Economists have estimated
that malaria costs sub-Saharan Africa at least $. billion a year.22 In fact,
. Full Report: World Health Report .
. D. S. Shepard, M. B. Ettling, U. Brinkmann, and R. Sauerborn. The Economic Cost
of Malaria in Africa, Tropical Medicine and Parasitology (): .
47 Global Health and the Scientific
Research Agenda

the total burden may be higher in that nations with endemic malaria
have economic growth slowed by . percent per year, suggesting that the
economy of Africa might be larger by a third had malaria been wiped out
two decades ago.23 Effective technology for eliminating malaria in Africa
might thus serve as a basis for ameliorating other forms of suffering.
Plainly, if such socio-economic considerations were incorporated into a
refined conception of a diseases fair share, they would only increase the
gap that divides fair share from actual expenditure.
Before we consider the possibility that the misalignment grounds an
obligation to change the ways that research is funded in the affluent
world (which is, after all, where research happens), well close this
section with a quick look at the ways that resources are actually divided.
The funding for science that comes out of biopharmaceutical business
roughly matches the funding provided by government. Each of the ten
largest drug companies has, in theory, the capacity to direct in excess of
a billion dollars a year to any biomedical research problem.24 In total, the
annual research and development expenditure of the international
pharmaceutical community is estimated at $ billion. Each year, the
lions share of drug companies resources is directed toward the most
profitable projects they can find, and the resulting research agenda is
dramatically imbalanced. Drug companies aspire to revenues of around
a billion dollars a year for a single drug. In the case of tuberculosis, for
example, theres no wealthy market for any new drug, so the profits are
unavailable, and its hardly surprising that there are no such drugs in the
pipeline of any major company. Over twelve hundred new drugs were
developed between and , but only thirteen were aimed specifi-
cally at tropical diseases.25
Plainly, the market imposes constraints on drug research; even in a
merciless market, however, some corporate idealism is possible. Merck
maintains a vaccine division (the only one in the major drug companies)
despite its lower profitability relative to other projects. In response to
political pressure, Merck also has also invested modest resources in
adapting a veterinary drug, Ivermectin, for combating a form of river
blindnesswhich infects million people in the world today and has

. J. L. Gallup and J. D. Sachs, The Economic Burden of Malaria, American Journal

of Tropical Medicine and Hygiene (): .
. Pfizer, currently the largest drug company, spends $ billion a year on research.
. See Silverstein, Millions for Viagra.
48 Philosophy & Public Affairs

blinded , of them, almost all in Africa.26 Merck donates the drug

to a project (led by WHO and the World Bank). Investors havent pun-
ished the company for these efforts, partly because theyre undertaken
on not too large a scale, and partly because they bring public relations
benefits. (The drug, Mectizan, features both in Mercks own promotional
material and in press coverage of the ethical stances of drug companies.)
Government funding of biomedical research is virtually entirely the
province of the NIH, which, as already noted, will distribute about $
billion in research funds in . The NIH clearly displays the / gap.
Out of the $ billion, $. billion are allocated to cancer research, $.
billion to the National Heart, Lung, and Blood Institute, $. billion to
the National Institute of Diabetes and Digestive and Kidney Diseases,
and $. billion to the National Institute of Neurological Disorders and
Strokes. In total, $. billion, or just over half, of the NIH budget goes to
departments that are clearly oriented toward solving only affluent-world
problems.27 The remaining funds go to departments that might be ori-
ented toward the diseases that afflict the poor, but, by and large, are not.
On the borderline between affluent-world and poor-world problems
is AIDS. Half of the NIHs $. billion AIDS budget in went toward
vaccine research (directly or indirectly); since then the NIH budget has
doubled, as has the amount given to AIDS research. As the emphasis of
AIDS research shifts further toward vaccine development, the NIH is
now investing significantly more than a billion dollars in trying to dis-
cover a vaccine for this disease. Of course, this reflects AIDS unusual
status as an infectious disease that is a dangerous killer both in the poor
parts of the world and in affluent nations.28
Compared to these figures, the amount of money explicitly set aside
for tropical disease is tiny. The only department officially devoted to
international health problems is the Fogarty International Center, whose

. WHO Factsheet: Onchocersiasis, World Health Organization, Geneva, February

.
. In addition to those mentioned, the other pertinent departments are: NIMH, NIA,
NIAMS, NIAAA, NIDCR, and NIDCD.
. As is well known, the plight of poor people who suffer from AIDS is exacerbated by
the inflated prices of the drugs used to combat HIV in the affluent world. We applaud the
Bush administrations initiative to commit funds to fighting AIDS in Africa, and hope that
this will be carried through, that the program will be extended to other impoverished parts
of the world, and that it will not be accompanied by prohibitions against the use of effec-
tive methods for controlling the spread of the disease (i.e., condom distribution).
49 Global Health and the Scientific
Research Agenda

budget of $ million is almost a rounding error (although one for which

we should be thankful). The NIH budget also includes $ million for
the Global Fund to Fight HIV/AIDS, Malaria, and Tuberculosis. The
examples dont exhaust the list of tropical disease projects, but they
are representative. Indeed, when one factors in the medical needs of
American military personnel who might have to serve in remote places,
its clear that government support of biomedical research only occa-
sionallyand accidentallyaddresses the needs of the poor.
What would happen if some of the funds currently earmarked for dis-
eases that afflict people in the affluent world were diverted to such dis-
eases as malaria and tuberculosis? As we have said, that is an empirical
question and we dont know the answer. We do, however, believe that it
would take some serious work to defend the idea that taking a few
million dollars from cancer research would materially lessen the chances
of making progress in fighting cancer.29
A final, prudential, point: some of the infectious diseases that
flourish in the non-affluent world have already become drug resistant
and have returned to haunt the rich. Tuberculosis is an obvious case: as
resistant strains reach the United States, the death rate from the disease
remains low, but the costs of controlling the infections has risen to
between $ million and $ billion.30 Many leading microbiologists
believe that its possible that, in seventy years, current trends may leave
us facing a world in which no antimicrobial drugs remain effective.31
Further, without more attention to the possible ways that increased
clearing of tropical rain forests encourages pathogen jumping from
nonhuman animals to our own species, we may be vulnerable to extra-
ordinarily devastating diseases.32 In short, its not clear that the division
of disease burdenwith the protected rich and the vulnerable pooris
sustainable.

. It should also be noted that the NIHs budget is currently growing at a rate enor-
mously in advance of inflation; in it will have doubled in five years, from $ to $
billion. Thus an increase to a half-billion dollars each in the annual funding for the malar-
ial and tuberculosis vaccine projects would do no more than slightly slow the rate of growth
of research into chronic diseases.
. Ruth E. Brown, Bess Miller, William R. Taylor, Cynthia Palmer, Lynn Bosco, Ray M.
Nicola, Jerry Zelinger, and Kit Simpson, Health Care Expenditures for Tuberculosis in the
United States, Archive for Internal Medicine (): .
. Garrett, Betrayal of Trust, p. .
. See Garrett, The Coming Plague.
50 Philosophy & Public Affairs

III
There are familiar arguments to the effect that we should do something
about the situation just reviewed, arguments that make the imperative
moral rather than merely a counsel of prudence. From a straightforward
utilitarian perspective its not hard to argue that frivolous expenditure by
affluent people should be redirected toward the alleviation of human
misery, and, in the same way, one can indict the investment in research
into new lifestyle drugs or the current neglect of many diseases.33 Even
without commitment to utilitarianism, or any form of consequentialism,
its hard to resist the conclusion that the imbalance of effort represented
by the / gap imposes some form of obligation to change how global
health is pursued.34 The amount of suffering that lies behind the recita-
tion of facts in Section IIcondensed and sparse though our review is
calls out for a response, and one cannot simply dismiss the burden of
disease as distant or no concern of ours.35 If pleas for attention to global
health (or global hunger) are met with apathy, that is surely the result of
a sense that the case is hopeless; that no matter what we might do, the
problem wouldnt be solved or even significantly alleviated.36
In our judgment, the sense of hopelessness arises from the conclu-
sions that are typically drawn from the moral exhortations: although
its correct to think that the skewed distribution of the worlds disease
burden requires somebody to do something, we think its important to
think broadly about what which individuals (or institutions) should do.
The standard arguments focus on ordinary citizens of affluent nations
and on possibilities for giving aid to distant people who are afflicted with

. The arguments are made with great force by Peter Singer, Famine, Affluence, and
Morality, Philosophy & Public Affairs, (): ; Peter Unger, Living High and
Letting Die (New York: Oxford University Press, ).
. Thus we wont try to offer a moral theory to buttress the claim that we have a prima
facie obligation to respond to the plight of afflicted people in poor countries, because we
dont see any plausible moral theory that fails to endorse any such obligation. The real
issue, we maintain, is whether the size and severity of the challenge undercuts the view
that what appears to be obligatory is genuinely demanded of us.
. Unger, Living High, makes the case against evasion on such grounds, and does so
with enormous thoroughness and clarity.
. Unger, Living High, refers to this as futility thinking, and we agree with him both
that such thinking lies behind the evasion of moral responsibility and that it is fallacious.
As well explain in the text, we think that the fallacy is a bit more complicated than Unger
takes it to be.
51 Global Health and the Scientific
Research Agenda

treatable diseases; philosophical concern for the health of the poor

typically doesnt focus on opportunities for research but rather on the
applications of existing knowledge that are not made because the funds
are lacking. In the classic example, were invited to consider two options,
responding to a request for oral rehydration therapy that will save the
lives of thirty children or failing to make that response.37 Realistic
predicaments would allow a far more extensive set of options, and its
worth noting that its not self-evident that relations among choices are
always preserved when the choice set is enlarged. We dont think that this
is more than a philosophical quibble in the case at hand; our aim is to
explore a much more systematic strategy for responding to the global
burden of disease, not to dispute what appears to us to be uncontrover-
sialto wit, that if our choices were restricted to sending the money or
not sending it wed have an obligation to do the former.
Numbing apathy sets in, we believe, because the lack of any system-
atic strategy makes plausible a skeptical response.38 This response has
three components: the first focusing on the size of the problem, the
second on the costs of addressing it, and the third on our obligations to
those who stand in special relationships to us.
The skeptic starts from the idea that sending the money to cover the
costs of oral rehydration therapy is simply applying a band-aid to a very
serious condition. Considered from one perspective the analogy is
clearly fallacious: the relievable suffering of an individual child doesnt
stand to the total affliction of people in poor countries as a minor cut
stands to the overall condition of a person with grave damage to major
organs and systems. The skeptical comparison, however, is intended to
draw attention to the systematic causes of suffering in impoverished
regions of the world, causes that are not in any way removed or muffled
by sporadic efforts to give oral rehydration therapy to a very few chil-
dren. The skeptic will point out that we may help a child who needs oral

. See Unger, Living High; Singer, Famine, Affluence, and Morality, uses similar
examples.
. This might be called the Malthusian Response, after the author of the Essay on
Population, with its attempt to vindicate the gospel dictum that the poor will always be
with us. The name is useful, because the standard diagnosis of Malthuss mistake views
him as overlooking the possibility that technological developments can disrupt his alge-
braic assumptions, but it also has misleading connotations, and so weve used the less spe-
cific skeptical response.
52 Philosophy & Public Affairs

rehydration today because she has had diarrhea from drinking contam-
inated water, but she will continue to live in a place where the water
supply is always untrustworthy, where the chances for many forms of
parasitic infestation are high, where malaria and tuberculosis are
endemic, where there is no chance for education, where women are
treated like commodities, where there are interminable tribal squabbles
that flare into hostility, and so on and so forth. Oral rehydration today
is a band-aid in that it fails to remove any of the many causes of her
continuing distress. Thus, it isnt the case that, all by itself, the action of
giving will genuinely reclaim the lives of those toward whom the aid is
directed.
One might dispute the point, arguing that the assessment that the
child saved today will be vulnerable tomorrow is unduly pessimistic.39
Yet the skeptic will remind us of the full range of the suffering reviewed
briefly in Section II. Every month, a million children suffer and die of the
same conditions we have attempted to alleviate, and for this vast major-
ity the world remains unchanged. To all appearances, when we and the
few children we have saved are gone, this tide of mortality will still roll
in unabated. Anybody studying it will see our efforts as a gesture, a tiny
island of generosity, not a solution.
Once we are aware of this point, we can appreciate a second skepti-
cal concern. To recognize the obligation to give now would produce a
situation so little different from the initial state that we would have an
equally pressing obligation to go on giving. The skeptic invites us to con-
sider two possible futures, the Status Quo and the Bleak World. In the
Status Quo, children in poor nations suffer and die somewhat earlier
than they do in the Bleak World; in the Bleak World their lives are pro-
longed, thanks to generous responses from affluent people. In the Bleak
World the entire population of the affluent world does what is viewed as
their obligationthey give and they keep givingbut the scale is so large
that the donors find themselves committed to a policy that withdraws
funds from many, indeed most, of the enterprises that give pleasure to
people in the affluent world: opera houses and theaters close, sports
tournaments are cancelled, ancient monuments are allowed to decay,

. Unger (Living High, pp. ) suggests that affordable contributions would give
a child in a poor country a percent chance of reaching the age of , so he would
presumably make the charge of pessimism. The skeptic might respond that the success
criterion that Unger uses is still too weak.
53 Global Health and the Scientific
Research Agenda

any natural preserves that could be turned to purposes of alleviating the

suffering of the poor are commandeered to such ends, and so on. The
benefits gained by this reallocation of resources are only slight: the lives
rescued from disease and malnutrition are somewhat longer and some-
what less painful. In the end, however, many things that seem to make
the actual world a better place are forsaken without completely solving
the problem of global disease.40
Theres an obvious reply: were not faced with a choice between the
Status Quo and the Bleak World. Perhaps we can make significant
improvements to the actual world by diverting funds from rich countries
to poor countries and relieving the burden of suffering there. As we shall
argue below, we think that this is correct, but we dont want to engage in
controversy about just how large the sacrifices from continued giving
would have to be; instead, we want to enlarge the space of options avail-
able. If our choices were restricted to individual giving in reaction to
individual affliction, then we think that its genuinely unclear how drastic
the consequences for life in the affluent world would be.41 Further, the
skeptic reminds us that it isnt evident that alleviation of suffering
trumps everything else, that there are no occasions on which it is better
to permit additional suffering in order that some other good be realized.
At this point, we can introduce the third theme in the skeptical
response. If the costs are as skeptics believe, then the task before us isnt
simply to give up a few luxuries in order to address suffering among the
distant poor, nor even to settle for a more Spartan public and cultural
life. There are sacrifices we must make in the provision of goods to those
with whom we stand in special relations: our children, our elderly
parents, our loved ones, our friends. Perhaps in some instances it will be

. Here the skeptic appeals to the kinds of considerations that lead Derek Parfit to
identify the claim that a world in which billions live lives that are just worth living is prefer-
able to a world in which a much smaller population enjoys lives that are full and rich as
the Repugnant Conclusion (Reasons and Persons, Oxford: Oxford University Press, ,
part IV). One of us has argued elsewhere that the problem posed by Parfit is insoluble; see
Philip Kitcher, Parfits Puzzle, Nos, (): . The upshot of that analysis is that
we ought to adopt a multidimensional approach to human well-being.
. Perhaps they would not be as dire as the skeptic believes. Surely (one might think)
if everyone in the affluent world gives to help the afflicted, the costs will be spread suffi-
ciently thinly that much of the quality of life can be sustained. We wont try to resolve this
issue because we believe that a purely reactive approach, even if effective, is not the best
way to address the problem.
54 Philosophy & Public Affairs

possible to justify expenditures on education on grounds that these will

make possible a more effective response to the enterprise of giving more
to the cause of global health, but all our domestic efforts to improve the
lives around us are subordinated to this more urgent project. There is a
genuine worry about what would become of our human relationships
if all our dealings with friends and relatives were held hostage to our
obligations to the many unknown people who suffer in distant places.
When we combine concerns about the lot of those who are dear to us
with the previous themes in the skeptical response, it is easy to antici-
pate a collective action problem. Assume that if all of us do our part, then
the burden of global suffering will be substantially eased; the skeptic
finds this assumption optimistic, but well allow it for the sake of argu-
ment. If none of us contributes, the distant millions will suffer and die.
You reason as follows: If all others contribute and you do not, then the
lives of faraway people will go almost as well (any difference will be
negligible), and those near and dear to you will benefit greatly from the
funds you would have sent (your elderly parents will enjoy the best care,
your children will have wonderful educational opportunities, and so
forth); if all (or even the vast majority) of the others fail to contribute and
you contribute, then the gains for the distant sufferers will be miniscule,
while your relatives and friends will be shortchanged. Not contributing
looks like a permissible course of action, given that it seems to have good
effects, whatever the others do.42
As we have said, we dont find the skeptical response conclusive, and
weve indicated some potential ways of contesting it. But we dont think
that the best way of meeting skepticism consists in pursuing those
optionsarguing, say, that a determined policy of individual contribu-
tion really can alleviate the global burden of disease. Instead we believe
that skepticism can help us think more intelligently about the problem
of distant suffering; reflections on scale direct us to consider root causes
and methods for eradicating them; attending to the various different
kinds of things we value can inform judgments about what sacrifices we
should struggle to avoid; noting that there is a looming problem of

. We wont try to set this up formally, but there are obvious similarities with both the
Prisoners Dilemma and the Tragedy of the Commons. A principal difference, of course, is
that the payoffs represent the values that a reflective, altruistic, moral agent would assign:
according to these payoffs, Dont Contribute is superior to Contribute no matter which
course of action the others pursue.
55 Global Health and the Scientific
Research Agenda

collective action can inspire ways of combining individual and collective

efforts.
Instead of a blanket obligation to send checks to UNICEF, citizens of
affluent nations have complex responsibilities that differ with their
opportunities and roles. Many of them should contribute more to reac-
tive efforts that aim to reduce the suffering of those already afflicted, but
many people have more specific obligations to contribute to agendas
that complement such projects. Some of the most pressing obligations
and most potentially potent contributions belong to members of the
biomedical research community.

IV
The skeptical trap is to frame debate by envisaging a line of worlds
ranging from the Status Quo to the Bleak World: the worlds are distin-
guished by the amount of money affluent people send to provide such
things as food, oral rehydration therapy, the drugs made available by past
research, and what we can expect by continuing with the same research
agenda. Well leave it to others to combat skepticism by claiming that the
scale along the line is wrong, or that even if its right, we have some oblig-
ation to head along the line that the skeptic draws. Our strategy is to get
off the line. As we noted in Section II, the creation and maintenance of
public health infrastructure could substitute efficient preventative mea-
sures for recurrent reactive treatments. Further, if the scientific research
agenda were modified (specifically, if far more resources were commit-
ted to studying the diseases that afflict people in poor nations) then we
might obtain systematic ways of eliminating major causes of suffering,
or at least diminishing their power.43 The social change we need to see
is the extension of successful public health infrastructure and practices
to parts of the world that lack them; the technological change is the cre-
ation of effective drugs and vaccines that can stop third world diseases
in third world conditions.
In our judgment, people in affluent countries have at least three dif-
ferent types of obligations. First, were obliged to provide relief for those

. In effect, were responding to skepticism in just the way that the historical Malthus
has been answered: we propose that seemingly inevitable human suffering can be evaded
by social and technological change.
56 Philosophy & Public Affairs

in imminent danger of severe disability and death; this involves precisely

the sorts of action that arguments about global suffering typically rec-
ommend, giving to organizations like UNICEF and OXFAM that respond
to crises. Additionally, the affluent nations collectively have an obliga-
tion to attack the causes that make these kinds of suffering chronic. This
obligation includes many projects, one of which is to bring to the poorer
parts of the world the basic preconditions for preventing disease and
dealing systematically with it: clean water supplies, clean air, shelter, a
reliable food supply, clinics and hospitals with electricity and facilities
for sterilizing equipment, schools.44 Citizens in rich nations have the
obligation to contribute taxes so that funds can be given to promote a
public health infrastructure in poor countries, and even to lobby their
governments to enact measures that increase taxation specifically to
provide the facilities weve listed. Third, the biomedical research com-
munity in the affluent world has the obligation to modify the current
research agenda so as to give much greater weight to investigations into
the diseases that produce extraordinary suffering among the poor. One
of the sacrifices that citizens of the affluent world should be prepared to
make is the diversion of resources from inquiries into chronic disease
that bring relatively small marginal chances of success. We note explic-
itly that the three-component program we envisage is directed at allevi-
ating the plight of the poor wherever they are; this means that one part
of the commitment to building public health infrastructure consists in
making public health facilities available to marginalized groups within
affluent societies.45
In the long run, building public health infrastructure is probably the
most important of these three goals, and it is the most difficult to
achieve. This is not, however, because such technical problems as secur-

. We include a system of education as a component of the public health infrastruc-

ture both because of the well-documented connection between educational level and con-
trolled birth rate, and because of the importance of understanding the social changes
sometimes needed to stop the spread of infection.
. Part of the obligation thus consists in making medical care available to all people
in the United States. For a clear and powerful description of the ways in which U.S.
medicine fails completely to meet the needs of many American citizens, see Garrett,
Betrayal of Trust, ch. . We also note that concern for the poorer members of affluent soci-
eties might also justify a policy of protecting those areas of biomedical research aimed at
diseases that disproportionately affect the poor; here we think of the impact that diabetes
and sickle-cell anemia have on African Americans.
57 Global Health and the Scientific
Research Agenda

ing a clean water supply and controlling mosquito populations are

unsolved, or because those solutions are prohibitively expensive. As we
discussed in Section II, the obstacles are political and social: even
though we know how to maintain the health of a population at a high
level, it can be extremely hard to do so in nations that are politically
unstable, or corrupt, or that distrust intrusion by the affluent powers.
Each region, each country presents very difficult special problems
of politics and culture that must be understood and energetically
addressed as their infrastructure is built.
We believe that the burdens of disability and disease are themselves
contributory causes to the poverty and political instability that make
good public health difficult to establish, and that the apparent indiffer-
ence of the wealthy nations to those burdens exacerbates these prob-
lems. For a sovereign state to allow outsiders to greatly alter the way it
handles public health requires a condition of trust that is frequently
lacking; one way to foster trust would be to show that the affluent world
has a serious concern for the suffering that infectious diseases bring to
the poorthat it is prepared to contribute emergency relief and, perhaps
more importantly, that it is prepared to devote the resources of its
sciences to the development of effective drugs and vaccines that will be
distributed to those who need them without regard for their ability to
pay. Since it is both a partial solution in its own right and the potential
key to making infrastructure improvement practical, a reformed health
research agenda is the third prong of our strategy, and perhaps the most
urgent one at this time.
In the rest of this article, well focus on this third prong of our
strategy. Well argue first that individual researchers have an obligation
to direct their research toward remedying the global research gap, and
subsequently that the institutions that fund and direct research have a
similar obligation.

V
There is a tendency to think of the obligations of scientists as very
simple; to view science as a private pursuit, driven by the love of truth.
To be sure, virtually everyone would concede that in seeking the truth
about aspects of nature, there are certain kinds of things that cant
be done: human subjects shouldnt be treated as mere means to the
58 Philosophy & Public Affairs

establishment of a decisive experiment, for example.46 Provided that sci-

entists conform to some relatively straightforward directives in design-
ing their experiments, however, a standard view is that they have no
further responsibilities; they should pursue the inquiries that strike them
as most promising.
In developing a more adequate picture of the role of the scientist and
the obligations it brings, its important to distinguish two different
settings in which scientific research is performed. Many people use
their scientific training to work on projects ultimately motivated and
constrained by the marketplace; in certain ways, these scientists subor-
dinate their role as scientists to their role as employees of profit-seeking
corporations; that isnt to say, of course, that they can legitimately seek
to defend findings they know to be false, or anything of a similar nature,
for in both roles they are subject to obligations of honesty. Well be prin-
cipally concerned with the second group of scientists, those whose
research isnt directly linked to their attempts to provide marketable
products; those for whom the stereotype of disinterested truth-seekers
is most plausible. So, in discussing the obligations of scientists, well be
concerned with those who work in universities or in laboratories with no
direct mandate from large corporations, not those whose research role
is framed by the quest for economic gains.47
We think that the minimal conception of the scientific role, sketched
in the first paragraph of this section, is incorrect. In becoming a scien-
tist someone takes on a new role, and that role brings obligations.48 To
fulfill the role, scientists should devote their energies toward achieving
the goals of the branch of science in which they work. What are these
goals? Not simply discovering truth, for the truths about the world are
too many and, for the vastly greatest part, too trivial. Were the biological
community to decide en masse that all further research efforts were to
be directed toward cataloging the exact number of bristles, hairs, or

. Two well-known examples of cases in which this moral requirement was violated
are the Tuskegee study of syphilis (in which Black patients were left untreated and igno-
rant of their condition) and the grotesque experiments of some Nazi doctors.
. Of course, the line between the university research laboratory and the marketplace
is becoming increasingly blurry. For reasons that will become obvious from our discussion,
we take this to be a matter of considerable concern; see also Philip Kitcher, Science, Truth,
and Democracy (New York: Oxford University Press, ).
. Here, and in what follows, weve been influenced by Michael Hardimons revival of
role obligations; see his Role Obligations, Journal of Philosophy (): .
59 Global Health and the Scientific
Research Agenda

similar structures found in individual organisms of an enormous variety

of species, wed rightly condemn the decision. This is because the aim of
the sciences is to answer significant questions, and theres no significant
question at which the enumeration of bristles is directed.
We suppose that there are two sources of scientific significance. Some
questions are significant because correct answers to them would enable
us to solve practical problems, to intervene in nature in some way, or to
predict its future course. Even investigations that have no direct practi-
cal benefits are sometimes undertaken because we suspect that there
will be possibilities of prediction and intervention that will ultimately
flow from them. Yet this is not all. Certain questions inspire scientific
effort not becauseor not merely becauseanswers would bring prac-
tical benefits (directly or indirectly) but because of prevalent human
curiosity. Its hard to imagine what pragmatic payoff would result from
sorting out the exact relationship among the australopithecines; even if
our understanding of the processes through which zygotes develop into
mature organisms would open up new possibilities for medical treat-
ment, that understanding is still valuable independently, for its own
sake.49
Well say that the pursuit of science in a society is well-ordered when
the research effort is efficiently directed toward the questions that are
most significant.50 In the contemporary world, the sciences are poten-
tially applicable to a large number of questions that impinge on the well-
being of many people, almost certainly too many for all of them to be
thoroughly and systematically pursued. Well-ordered science must
inevitably be selective, and we propose that it should respond to the
choices that would be made by an ideally informed collection of repre-
sentatives of divergent points of view, with each representative commit-
ted to learning about and taking seriously the needs and interests of each
of the others. To suppose that the collection of representatives be con-
fined to the points of view within the affluent world is, in our judgment,
to make an important moral error. Hence, our conception of well-
ordered science supposes that weight must be given to the needs of
those who suffer the enormous burden of infectious disease in the poor
regions of the world.
. Here we summarize a line of argument developed at much greater length in Kitcher,
Science, Truth, and Democracy.
. This conception is articulated further in ibid., ch. .
60 Philosophy & Public Affairs

Our defense of the claim that a narrower conception of the set of view-
points represented is morally mistaken will be relatively brief. Its a
matter of historical accident that some people live in societies with the
resources to commit to scientific research, and, we suggest, no ideal for
the direction of scientific inquiry should reflect such accidents. Its easy
to extend a familiar Rawlsian thought experiment and to imagine our-
selves choosing an ideal for the conduct of science without yet knowing
whether we shall find ourselves in one of the societies lucky enough
to support research; as we understand that thought experiment, the
obvious decision is to include the perspectives of all in the framing of
well-ordered science. Further, if, as we believe, the principal opposition
to responding to human needs, even when they are distant, is that such
responses are inevitably futile, then no such counterargument is avail-
able when the envisaged responses feature the development of the
sciences. The institution of science fulfills a valuable social function
because scientific knowledge brings us both intellectual and practical
goods, and societies invest in and support the institution largely because
they believe in the efficacy of research to address practical problems.51
Neither the claim that scientific inquiry is futile nor that it can only
address the needs of a restricted group of people has the least plausibil-
ity. Any narrowing of the class of viewpoints represented in well-ordered
science thus reflects not hard-headed realism but a callous neglect of
those who are poor and distant.
Having outlined and briefly defended an ideal, we now consider what
obligations flow from it. Taking on the role of a scientist brings with it
the responsibility of contributing to well-ordered science. That respon-
sibility doesnt necessarily preclude the possibility of addressing issues
that have purely theoretical significance, of attempting to satisfy human
curiosity. Yet it would be a travesty of well-ordered science to propose
that only two kinds of inquiries be pursued: those that attempt to satisfy
disinterested curiosity and those that seek to meet the practical needs of
citizens of affluent nations. Hence, in the presence of the / gap,
theres ample reason to think that scientific research is not promoting its
proper goal. In consequence, scientists have an obligation to do what
they can to remedy the situation.

. For further development of this theme about the function of scientific research, see
Philip Kitcher, The Scientists Role (John Wesley Powell Lecture, ).
61 Global Health and the Scientific
Research Agenda

For some scientists, of course, theres little that can be done directly;
they work in fields that are too remote from those that bear on the
neglected issues. Biomedical researchers, however, can often do much
more. Some have the option of pursuing inquiries that might have
value in relieving arthritic symptoms or of working on some model
organism that might be useful in studying infectious disease. By taking
the latter course, they can move the research community closer to a state
of well-ordered science. In our judgment, they have the obligation to
do so.
Further, all scientists have an obligation to engage in political
activism, to campaign publicly for greater investment in research that
would address the disease burden of the poorer regions of the world. But
the capacity of ordinary scientists to bring about much change is prob-
ably limited; they are typically inclined and trained to discover facts, not
influence political opinion. The administrators who control the flow of
cash to science play on a more public stage. We turn next to their
responsibilities.
Several different types of organizations fund enough research sub-
stantially to affect science policy. Governments of rich nations are obvi-
ously important, although we think it more useful to consider these as a
few different, semiautonomous institutionsin the United States, for
example, we might list the NIH, the NSF, and the complex of national
laboratories. A second type consists of the biopharmaceutical industry,
considered either as a bloc or as individual companies. Independent
foundations like the Wellcome Trust and the Gates Foundation make up
a third. The university research system is a fourth, although once one
considers the sources of funding it becomes clear that university labo-
ratories cant operate independently of the government institutions and
a few foundations. Were going to concentrate on the American research
effort, and well reduce the four categories to two: the first is the gov-
ernment, primarily in the form of the NIH and the university labs it
funds; the second is the biopharmaceutical industry.
If, as weve claimed, the / gap represents a departure from
well-ordered science, then while we may view scientists as having been
insufficiently vocal in protesting, the root of the trouble seems to be the
research priorities set by the NIH and the biopharmaceutical companies.
Well deal with the two sources of funding separately, beginning with the
case of publicly funded science.
62 Philosophy & Public Affairs

As noted in Section II, the priorities of the NIH budget are badly
misaligned with the global burden of disease. In light of the promise
of technological solutions to problems that cause massive human
sufferingsuch as pathogen sequencing, with consequent attempts to
construct vaccinesNIH administrators have a moral obligation to do
what they can, within the budgetary constraints imposed, to modify the
current grossly skewed distribution; since the constraints make it impos-
sible even to approximate a distribution that would make research on a
disease comparable to the toll exacted by that disease, they have the
broader responsibility to protest the legislative guidelines. By the same
token, legislators ought to repeal the existing constraints in ways that
would make it possible to move much closer to well-ordered science.
Citizens have the obligation not only to write their individual checks,
but also to support these legislative changes.
How might these claims about moral responsibility be evaded? If we
are right in claiming that the principal source of resistance to the origi-
nal arguments enjoining individual contributions to disease relief lies in
the sense that such contributions are futile, then the crucial issue is
whether our revised moral imperative faces a similar skeptical response.
Skeptics can certainly point out that changing the research agenda isnt
guaranteed to bring success, either in addressing any particular disease
or in creating a climate in which public health interventions would be
welcomed. But, of course, we typically lack guarantees when we invest
in biomedical research, and, as we noted in Section II, theres good
reason to anticipate results from an ambitious vaccine development
project. If the calculated risk of investing in research pays off, the scale
of the potential reward is inspiring. For example, AIDS, tuberculosis, and
malaria are each single causes of a million or more deaths a year, along
with considerable related suffering. If we can develop and distribute an
effective vaccine for just one of these diseases, the gain to human well-
being is hard to express adequately. Since our hope here is to eradicate
or completely control the disease, the skeptics worry that were com-
mitted to an indefinite sequence of draining contributions is no longer
germane. Since the costs of such vaccine research are low on a global
scale, and returns to world security and the world economy great, we
need not worry that it will drive us toward the Bleak World or make us
do poorly by our commitments. Continued AIDS vaccine research and
credible attempts at malaria and tuberculosis vaccines are the least futile
of enterprises.
63 Global Health and the Scientific
Research Agenda

Suppose, however, that serious investments in research on the dis-

eases that afflict poor nations dont succeed. To what extent should we
pour more funds into the effort? We think that this question is hard to
answer responsibly in advance. Judgments of the promise of research
strategies are rightly revised as investigators discover new things, and it
might turn out that the promise we have envisaged proves unreal. Our
judgment is that given the prospects as they now appear, a commitment
to a distribution of research that assigns each disease its fair share is
morally required. That, by itself, doesnt fix the level of total funding for
biomedical research. We are inclined to believe that affluent nations (in
particular the United States) could afford to increase the research budget
so that such diseases as cancer were still funded at approximately their
current levels (although this would depend on serious exploration of the
likely returns from investment at the margin), and the entire research
budget were constructed by indexing to these amounts by applying the
fair share principle.52 We note also that a public commitment to this
investment would be a clear declaration of concern for the suffering of
people in poor countries, and that undertaking the commitment might
well create an atmosphere of trust that would allow the exportation of
first world public health infrastructure, even if the technological solu-
tions were less frequent than we believe likely.53
As we noted in Section II, the funding for science that comes out of
biopharmaceutical business roughly matches that provided by govern-
ment, and well turn now to the question of whether pharmaceutical
companies have any obligations to change their ways. Theres an obvious
defense of Big Pharmaits business, after all, is business. The demands
of the free market make it impossible for companies to devote more than
token resources to unprofitable projects. Imagine that a company were
to seize the moral high ground, turning a significant proportion of its
research resources toward vaccines for tropical disease. Profits would
fall, investors would go elsewhere, and the decline might be so rapid that

. For a more detailed account of how levels of funding might be set under well-
ordered science, see Kitcher, Science, Truth, and Democracy, ch. . We note that the level
we envisage here seems not to be vulnerable to the skeptical concerns about the sacrifice
of many things that are valued in the affluent world.
. Given our account of the roles of citizens, of legislators, and of NIH administrators,
its relatively easy to see that any worries about the potential collective action problem are
forestalled; in effect, were envisaging a classical solution to the difficulties of coordina-
tionto wit, the use of government as a coordinating mechanism.
64 Philosophy & Public Affairs

the research would never be completed. Plainly the market constrains

action.
Yet the most obvious point to make about the pressures against devel-
oping vaccines for understudied diseases is that we neednt take the
forces of the market as an unmodifiable given. Even the most devout free
marketer should recognize the legitimacy of questions as to whether a
given domain is best organized by subjecting it to the organization of the
market; its worth bearing in mind that Adam Smith thought it impor-
tant to argue that international trade should be governed by the market,
and that he believed that there were preconditions for the operation of
free markets that were best provided in other ways.54 If half the funding
for scientific research is allocated by a process that muffles (or silences)
the major concerns of percent of human beings, and if those concerns
are not being adequately addressed, then theres a prima facie case for
thinking that the market organization of biopharmaceutical research
ought to give way to a different system. One obvious possibility is for
government to provide incentives to companies that invest in develop-
ing drugs for underresearched diseases, perhaps by allowing those com-
panies tax relief. To defend the status quo one would need to show that
these, and other ways of trying to generate a less skewed distribution of
research effort, involve sufficiently large costs in efficiency.55 Until the
argument has been given, theres a moral obligation to modify the
agenda of biopharmaceutical research to provide much greater funding
for vaccine projects (and similar programs aimed at reducing the disease
burden on the poor).
The line of argument we have broached here leads into more general
issues about the role of the market in issues of health and health care.
There are serious questions about whether we should honor the alleged
efficiency of the free market when it leads to skewed (often grossly
skewed) distributions in the availability of health insurance and in the
relative density of doctors, clinics, and hospitals in different places, as
well as to inflated costs for drugs and other forms of treatment. We shall

. See Wealth of Nations, books and . The cluster of issues about the proper sphere
of governmental provision continues to occupy political economy after Smith; see, for
example, J. S. Mills Principles of Political Economy.
. One intriguing possibility, suggested to us by the Editors of Philosophy & Public
Affairs, is that scientists form nonprofit organizations in which to pursue biomedical
research. This seems to us an option well worth exploring.
65 Global Health and the Scientific
Research Agenda

rest content with adding the possibility that pharmaceutical companies

might be required to devote a portion of their research investment to
alleviating the global disease burden to a family of proposals that favor
tempering the relentless pursuit of efficiency: proposals for universal
health care coverage, for a more equitable distribution of medical access,
and for the limitation or abolition of patents. Detailed examination of
how the current practices of pharmaceutical companies could be mod-
ified to accord better with the obligations we have been highlighting
must await a future occasion.
Our current biomedical science is a very long way from being well-
ordered. Collectively, we have the obligation to do much better. Weve
argued that the obligation takes different forms, depending on the role
(or roles) that a person plays in the direction of scientific research.
Legislators have the responsibility to modify the conditions under
which biopharmaceutical companies and NIH administrators allocate
research resources; drug company executives and administrators should
do whatever legislators cannot to correct the imbalances in the distrib-
ution of research effort; individual scientists should reorient their
research, where they can, in ways that bear on the diseases that afflict
the poor; and ordinary citizens ought to support all these effortseven
if it means that some of our personal interests are thereby sacrificed.