Essays

Whatever Happened to Human
Experimentation?
b y CAR L ELLIOT T

S

everal years ago, the University of Minnesota hosted a
lecture by Alan Milstein, a Philadelphia attorney specializing in clinical trial litigation. Milstein, who does
not mince words, insisted on calling research studies “experiments.” “Don’t call it a study,” Milstein said. “Don’t call
it a clinical trial. Call it what it is. It’s an experiment.”
Milstein’s comments made me wonder: when was the
last time I heard an ongoing research study described as a
“human experiment”? The phrase is now almost always associated with abuses. Ask people what they think of when
they hear the phrase “human experimentation” and chances
are they will reply, “Nazis.” James Jones gave his book Bad
Blood  the subtitle The Tuskegee Syphilis Experiment. Philip
Zimbardo’s controversial study on the psychology of imprisonment is commonly referred to as “The Stanford
Prison Experiment.” New Zealand’s most notorious research scandal, on women with cervical carcinoma in situ at
Auckland Women’s Hospital, is known colloquially as “The
Unfortunate Experiment.” Asking a prospective subject to
sign up for a medical experiment would probably get roughly the same response as asking him or her to sign up for a
police interrogation.
It wasn’t always this way. In the early days of American
bioethics, scholars used the word “experimentation”
in the same neutral way that they later began to use “research study” and “clinical trial.” Hans Jonas titled his famous 1969 Daedalus  article “Philosophical Reflections on
Experimenting with Human Subjects.” In the 1966 essay
“Ethics and Clinical Research,” Henry Beecher made frequent use of phrases such as “experimental subjects,” “human experimentation,” and “experimentation in man.”
Carl Elliott, “Whatever Happened to Human Experimentation?,” Hastings
Center Report 46 (2016): 1-4. DOI: 10.1002/hast.531

Richard McCormick and Paul Ramsey often used the
phrase “experimentation” in their debate in the mid-1970s
about research on children. This usage reflected the vocabulary of medicine itself, in which the word “experiment”
had not yet acquired a menacing undertone. For instance,
Ancel Keys did not hesitate to use “experiment” in describing his work on the physiological effects of semistarvation
at the University of Minnesota in the 1940s. In the twovolume book that resulted in 1950, The Biology of Human
Starvation, Keys and his colleagues even titled one chapter
“The Minnesota Experiment.”1
By the time I began working in bioethics in the early
1990s, however, “experiment” was being phased out. Today,
when anyone involved in health care uses “experiment,” it is
usually in reference to scandals from the past. Over the past
ten years, for instance, PubMed lists only thirty-four articles
with the phrase “human experimentation” in their titles, and
thirty of those concern either historical research abuses or
ethically controversial research.
Experimentation has also disappeared from codes of ethics. When the Nuremburg Code was published in 1948, it
referred to “experiments,” “experimental subjects,” and “experimental physicians.” It made no mention whatsoever of
“research.” The first version of the Declaration of Helsinki,
issued in 1964, employed a combination of phrases: sometimes “experimentation,” sometimes “research study.” That
mixed language stayed relatively consistent through the
next six revisions of the Declaration. But in 2008, when the
Declaration was revised in Seoul, “experiment” was almost
completely purged from the document. The word is used
only once in the current version, and then only in reference
to the need for “animal experimentation.”
What is the reason for such a dramatic shift? A charitable answer is that it simply reflects greater attention to linH AS TI N GS C E N TE R REPORT

1

guistic accuracy and precision. Not all research studies are
experiments. Nor are they all clinical trials. Chart reviews,
surveys, ethnographies, epidemiological studies, and many
other kinds of medical research could hardly be described
as “experiments,” insofar as “experiments” suggests scientific
testing conducted under controlled conditions. In addition,
as the authors of The Belmont Report pointed out, not all
experimental procedures count as research. A radical new
procedure may be untested and thus experimental but not
part of a formal research study. Maybe the phrase “research
study” is simply intended to be broader and more inclusive.
But today, even the most obviously experimental research
studies are described in language that seems intended to hide
that fact. Take, for instance, the disastrous trial of TGN1412
at Northwick Park Hospital in 2006.2 This was a phase I
study of a monoclonal antibody that, within minutes after it
was administered, sent six subjects on the way to multisystem organ failure. If any drug could qualify as experimental,
it was TGN1412, a novel immune-modulatory compound
that had never before been administered to humans. Yet in
the consent form that the research subjects were asked to
sign, the words “experiment” and “experimental” never appear.3
It’s not hard to understand why. In 1995 the Advisory
Committee on Human Radiation Experiments surveyed
patients to compare their perceptions of the terms “clinical trial,” “clinical investigation,” “medical study,” “medical
research,” and “medical experiment.” Patients ranked the
phrase “medical study” as by far the most benign-sounding
term. A “study” means looking at records, said one patient.
“It is mostly paperwork, documents, or the books and
things.” Patients thought the phrase “medical experiment”
suggested a riskier, scarier undertaking. One said, “I envision all kinds of weird things done to the body, and I assume
that’s not true, but also I envision a medical experiment
maybe . . . done in a laboratory sealed up somewhere where
no one even knows what [is] going on.” Some people said
they would not consider taking part in an “experiment” unless they were terminally ill.4
These responses point to a more plausible explanation for
the shift away from “experiment.” Once the word became
associated in the public mind with abuse and exploitation,
calling something an “experiment” started to sound like an
ethical condemnation. Today the choice whether to refer to
the testing of a new drug in humans as an “experiment” or
a “research study” seems like an ideological decision, like
whether to call waterboarding “torture” or “enhanced interrogation” or whether to use the term “alien” instead of
“immigrant.” Many writers and organizations, attempting
to remain neutral, simply choose the language sanctioned
by the authorities.
In this case, it is clear what the authorities prefer. The research establishment has a vested interest in using language
that portrays its work as safe, carefully regulated, and socially beneficial. And of course, if the language itself does not
comply, it can be pushed. In 2008, for instance, in order to
2 HASTINGS CEN T E R R E P ORT

combat unfavorable public opinion about medical research,
Eli Lilly funded a public relations campaign called “Medical
Heroes in Everyday Places.” Developed by the public relations firm Ogilvy and Mather and organized through the
Center for Information and Study on Clinical Research,
the purpose of the campaign was “transforming public perception of the clinical trial participant from ‘guinea pig’ to
‘medical hero.’”5 An updated campaign in 2015, titled the
“Rethink Research Competition,” produced slogans such
as “Heroes aren’t hard to come by,” “Remedi THIS,” and
“Going study.”
Perhaps it should not be surprising that bioethicists have
changed their language as well. Today’s bioethicists are employed by hospitals and medical schools, publish in medical
journals, and apply for NIH research grants. Some consult
for pharmaceutical companies and contract research organizations. In this world, medical language serves as a kind
of code for insiders. The choice to use a particular word or
phrase indicates where the user is situated in relation to the
invisible lines between researchers and subjects, clinicians
and patients, advisors and critics. As long as bioethicists
worship in the same church as the scientists and the clinicians, it will feel natural for them to sing the same hymns.
The choice to abandon the word “experiment” is emblematic of a larger movement beginning in the 1990s, in
which many bioethics scholars moved from being critics of
the research enterprise to being its champions. Probably the
most striking example is the argument now made by a number of prominent bioethicists that every citizen has a moral
duty to enroll in research studies.6 “There is a prima facie
obligation to participate in biomedical research,” a group
from the Department of Bioethics at the National Institutes
of Health has written—an obligation that includes even research sponsored and designed by the pharmaceutical industry. “Whether a study is publicly or privately financed,” the
group insists, “has little bearing on the obligation to participate in that study.”7
Theirs is still a minority view. But a much larger and more
influential shift can be seen in the way many bioethicists have
come to champion the social benefits of research, especially
to particular “communities.” In the 1990s, for instance,
feminist writers began to argue successfully that excluding
women from clinical trials meant excluding them from the
social benefits of medical research—namely, knowledge
about how illnesses affect women and what drugs and procedures might be best used to treat them.8 Conceptualized
in this way, research guidelines did not simply protect subjects from harm; they denied communities access to health
care. Once research was described anew as a social benefit, it
could be used as a justification for recruiting subjects from
some of the very populations that had been victimized in the
past. Only twenty-five years after the Tuskegee syphilis experiment, for example, the authors of Beyond Consent would
suggest that “overprotection” of racial minorities may result
in a failure to address their health needs.9

Along with the reconfiguration of research as a social
benefit have come efforts to replace the term “research subject” with the softer, friendlier “research participant.” In a
2001 report, the National Bioethics Advisory Commission
laid out its rationale for the change. The word “subject,”
the commission said, “connotes society’s worst fears about
research participation, rather than its aspirations.” One way
to dispel such fears is to use the phrase “participant” in order
to emphasize that “individuals should be active, not passive,
in the decision to enroll in research studies.”10
This particular change was driven at least partly by patient advocates. During the early days of HIV-AIDS activism, many advocates resented being seen as mere test
subjects and wanted to play a more active role in planning a
research agenda. To many, the phrase “research participant”
felt empowering. Whether subjects were actually empowered
is another matter. This change to the language may make
subjects sound more like collaborators in the research enterprise, but it does not change the fact that the subjects are the
ones taking the experimental drugs—rather than the ones
designing the protocol, administering the medication, analyzing the data, or authoring the publications. A handful of
activists understood this power relationship more clearly. In
the mid-1990s, when Bob Helms began publishing Guinea
Pig Zero, his zine for research subjects, Helms and his fellow operatives used phrases such as “lab rat,” “guinea pig,”
“brain slut,” and “medical meat puppet” in order to highlight their relative lack of power.11
Because no agency tracks deaths and injuries in medical
research, it is hard to know whether research has become safer or more dangerous in recent decades. But research scandals are not hard to find. The most notorious recent scandal,
of course, is the death of Jesse Gelsinger at the University of
Pennsylvania in 1999.12 But there have been many others.
There were the deaths of healthy volunteers such as Nicole
Wan at the University of Rochester, Ellen Roche at Johns
Hopkins, and Traci Johnson at Eli Lilly Laboratories.13 There
was Protocol 126 at the Fred Hutchinson Cancer Center, in
which at least twenty subjects died; Pfizer’s notorious Trovan
trial, in which eleven Nigerian children died; and the “bacteria-in-brain” episode at the University of California, Davis,
in which three neurosurgical patients died.14 My own institution, the University of Minnesota, has been involved in
a series of psychiatric research scandals dating back to the
1990s—most notoriously, the suicide of Dan Markingson
in an industry-funded antipsychotic study in 2004.15 The
poor and disenfranchised have been hit especially hard—for
instance, in the SFBC International episode in Miami, in
which researchers paid undocumented immigrants to test
the safety of unapproved drugs in a converted motel, or by
the ongoing recruitment of mentally ill subjects from homeless shelters in Philadelphia.16
It is entirely understandable that careful, conscientious
medical researchers would want to distance their work from
the exploitation and injustice of the past. And if the word
“experiment” conjures up past scandals, it is natural that

they would prefer to avoid it. But is that strategy intellectually honest? As early as 1996, George Annas argued that it
was not. “[F]or many potential research subjects, deception
or self-deception is inherent in our current research endeavors,” Annas wrote,17 comparing the language of clinical research to the political sloganeering of Orwell’s 1984. “Even
a cursory examination of modern human experimentation
demonstrates,” he said, “the pervasiveness of three doublespeak concepts: experimentation is treatment, researchers
are physicians, and subjects are patients” (p. 300).
It is hard to disagree with Annas. Yet it is even harder
to fight against the linguistic current. The shifts that Annas
warned about in 1996 are now so fully entrenched that
they strike no one as unusual, much less controversial.
Even Annas himself occasionally used the phrase “medical
research,” rather than “experimentation,” in his critique. I
have done the same here, of course. Doing otherwise would
strike many readers as an unnecessary provocation. To the
extent that this is true, the fight has already been lost.
Alan Milstein says that when he is deposing clinical investigators, he often asks them why they don’t use the word
“experiment” when they talk to subjects. Inevitably, they say
that the word carries too much baggage. To which Milstein
replies, “That is what is so wonderful about language. Words
carry meaning and history.”
1. A. Keys et al., The Biology of Human Starvation, 2 vols.
(Minneapolis: University of Minnesota Press, 1950).
2. E. Rosenthal, “When Drug Trials Go Terribly Wrong,” New York
Times, April 7, 2006.
3. Parexel International, “Informed Consent Form” (protocol number: TGN1412-HV), accessed October 23, 2015, http://www.circare.
org/foia5/tgn1412_consentform.pdf.
4. Advisory Committee on Human Radiation Experiments, Final
Report of the Advisory Committee on Human Radiation Experiments
(Washington, D.C.: U.S. Government Printing Office, 1995), 734.
5. D. Borfitz, “The Recruitment Problem,” Bio-IT World,
September 5, 2008, http://www.bio-itworld.com/issues/2008/
sept/patient-recruitment.html; B. Herskovits, “Returning the
Favor,”  Pharmaceutical Executive, January 23, 2007, at http://www.
pharmexec.com/node/237787.
6. J. Harris, “Scientific Research Is a Moral Duty,” Journal of
Medical Ethics 31 (2005): 242-8; R. Rhodes, “Rethinking Research
Ethics,” American Journal of Bioethics, 5, no. 1 (2005): 7-28; H. M.
Evans, “Should Patients Be Allowed to Veto Their Participation in
Clinical Research?,” Journal of Medical Ethics 30 (2004): 198-203.
7. G. Schaefer, E. Emmanuel, and A. Wertheimer, “The Obligation
to Participate in Biomedical Research,” Journal of the American Medical
Association 302, no. 1 (2009): 67-72.
8. R. Dresser, “Wanted: Single White Male for Medical
Research,”  Hastings Center Report 22, no. 1 (1992): 24-29; D. A.
DeBruin, “Justice and the Inclusion of Women in Clinical Studies:
An Argument for Further Reform,” Kennedy Institute of Ethics Journal
4, no. 2 (1994): 117-46; S. Sherwin, “Women in Clinical Studies:
A Feminist View,” Cambridge Quarterly of Healthcare Ethics 3, no. 4
(1994): 533-8.
9. J. Kahn, A. Mastroianni, and J. Sugarman, eds., Beyond Consent:
Seeking Justice in Research (New York: Oxford University Press, 1998),
6.
10. National Bioethics Advisory Commission, Ethical and Policy
Issues in Research Involving Human Participants, vol. 1, Report and

H AS TI N GS C E N TE R REPORT

3

Recommendations of the National Bioethics Advisory Commission
(Rockville, MD: NBAC, 2001), 16 (footnote 1), 33.
11. B. Helms, ed., Guinea Pig Zero: An Anthology of the Journal for
Human Research Subjects (New Orleans: Garrett County Press, 2005).
12. S. Stolberg, “The Biotech Death of Jesse Gelsinger,” New York
Times, November 28, 1999.
13. E. Rosenthal, “New York Seeks to Tighten Rules on Medical
Research,” New York Times, September 27, 1996; G, Kolata, “Johns
Hopkins Admits Fault in Fatal Experiment,” New York Times, July 17
2001; J. Lenzer, “Drug Secrets: What the FDA Isn’t Telling,” Slate,
September 27, 2005.
14. D. Wilson and D. Heath, “Uninformed Consent: What Patients
at the Hutch Weren’t Told about the Experiments in Which They
Died” (five-part series), Seattle Times, March 11-15, 2001, http://
seattletimes.com/uninformed_consent/; J. Stephens, “Where Profits
and Lives Hang in Balance,” Washington Post, December 17, 2000;
M. Lundstrom, “UC Davis Surgeons Resign after Bacteria-in-Brain
Dispute,” Sacramento Bee, August 25, 2013.

4 HASTINGS CEN T E R R E P ORT

15. C. Elliott, “Making a Killing,” Mother Jones, August-September
2010, 55-63; J. Rigert and M. Lerner, “University Kept Silent for
4 Years on Research Misconduct by Garfinkel,” Minneapolis Star
Tribune, August 12, 1993; M. Lerner and J. Rigert, “Professor
Accused of Coercing Hmong into U Drug Study,” Minneapolis Star
Tribune, October 10, 1993; J. Olson, “U Doctor Scrutinized over
Drug Research,” St. Paul Pioneer Press, March 19, 2009.
16. D. Evans, M. Smith, and L. Willen, “Big Pharma’s Shameful
Secret,”  Bloomberg Markets, December 2005, pp. 37-62; C.
Elliott, “The Best-Selling, Billion Dollar Pills Tested on Homeless
People,”  Matter, July 27, 2014, at https://medium.com/matter/didbig-pharma-test-your-meds-on-homeless-people-a6d8d3fc7dfe.
17. “G. Annas, “Questing for Grails: Duplicity, Betrayal and SelfDeception in Postmodern Medical Research,” Journal of Contemporary
Health Law and Policy 12, no. 2 (1996): 297-324, at 321.