Selected examples from history are discussed to illustrate the many
difficulties in judging scientific behavior. Scientific misconduct is not an a
priori given concept but must first be defined. The definitions of scientific
misconduct used in the USA and in Denmark are discussed as examples.
Denmark took a lead in the handling of
scientific misconduct by establishing the Danish Committees on Scientific
Dishonesty in 1998. As a political consequence of the committees handling of
the Lomborg case, the Danish legislature passed a new law for the committees
which came into force from 1 August 2005. The new law has restricted the work
of the committees. It remains to be seen
whether the Danish model is still a good solution for unveiling scientific
fraud. The role of science journalism in unveiling fraud in the sciences is
also discussed.
Good
Conduct in the Sciences
A platform for serious scientific misconduct
of an individual can develop when smaller irregularities are ignored by close
colleagues and management. You may forget to report properly about the origins
to your new idea or experiment, or you might give insufficient acknowledgement
to other peoples work, or borrow a couple of sentences without quotation. If a
researcher never has been introduced to basic rules for good scientific
practice by responsible colleagues he may end up finding such minor offences
natural and acceptable.
A state has laws for regulating the behaviour
of its inhabitants, not necessarily with the intention of punishing people but
in order to prevent undesired behaviour. Nevertheless crime – breaking the laws
- does take place, and it is important for a civilized society to have a legal
system to deal with crime in a proper way. In the same way, it is necessary
with a legal system to handle scientific misconduct. It is too late to start
thinking about misconduct when it appears, and it is too naïve to deny its
existence.
Examples
from History
Improper scientific behaviour is not a new
phenomenon. History is rich of examples of scientists who acted outside the
accepted standards of their profession [1].
The lesson from history that I hope to convey
by the following examples is that it is a very difficult task to find agreeable
and sensible attitudes to deal with misconduct in the sciences, but also that
it is a very necessary one.
Was
Darwin guilty of Misconduct in his ‘Origin of Species’?
It is now and then asserted that the famous
British naturalist Charles Darwin behaved improperly in his path breaking
publication of 1859 ‘On the Origin of Species by Means of Natural Selection’.
The assertion is that Darwin found his key mechanism, the concept of ‘natural
selection’, only when his countryman, the likewise eminent naturalist Alfred
Russel Wallace, sent Darwin a letter describing a theory about ‘biological
evolution’, and that Darwin did not give sufficiently credit to Wallace for
this. It is true that Darwin promptly sought to publicize the idea of ‘natural
selection’ and that he arranged by himself a joint paper with Wallace of 1858,
before publishing his own main publication in 1859. However, it should be said
to the excuse of Darwin that he had been conducting his own research for twenty
years and that his child died as he received Wallace’s letter from Indonesia,
where he stayed at the time. Furthermore, Wallace always later referred to the
concept of ‘natural selection’ as Darwin’s theory. There seems to have been no
bad feelings between the two men and hence there are no reasons to think of
Darwin’s behaviour in this matter as a case of scientific misconduct.
The discussion whether Darwin or Wallace
should have the priority to the principle of ‘natural selection’ in the
biological sciences is an example of how
misinformation over time can build up and support the attitude often found among
scientists that “we are all sinners, even the greatest scientists”. In this
connection it is interesting to note that more than 5 percent of American
scientists answering a confidential questionnaire admitted to minor degrees of
misconduct, according to an article in Nature [2] reporting that up to thirty
percent of the scientists admitted to have acted at odds with good scientific
practice.
There are good reasons to be on guard when a
colleague shows signs of bad manners in the sciences. Senior scientists should
consider it their duty to react on the spot - even if it is unpleasant - in
order to prevent that it may later escalate into proper scientific misconduct.
The
Discovery of the Double-Helix Structure of DNA
It is likely that the British molecular
biologist Rosalind Elsie Franklin was the true discoverer of the double-helix
structure of the DNA molecule for which her countrymen Francis Crick and
Maurice Wilkins together with the American geneticist and biophysicist James
Watson received a Nobel Prize in 1962, four years after Franklin’s death from
cancer at age 37. Franklin was employed by Wilkins to work on the project and
developed the X-Ray crystallography necessary to take the famous photograph. It
sat in her desk drawer until Wilkins informed Watson and Crick who removed it
from the drawer, claimed it as their own and went public. Is this a case of
fraud, plagiarism or the invisibility of women in science?
Inventions
and the Sciences
The invention of the telephone is a famous
example of seeming misconduct outside proper science as such [3]. Philip Reis,
a German professor and inventor, developed the apparatus and demonstrated it to
people. The American inventor Alexander Graham Bell appears to have witnessed a
demonstration and rushed on to develop his own apparatus and to patent the idea
on February 14, 1876. Bell had his patent registered just a few hours before
Elisha Gray, another American inventor. Elisha Gray and others sought to contest his priority but Bell was very
shrewd, investing in a powerful team of lawyers to frighten the
opposition.
Improper appropriation of intellectual
properties rightly belonging to others is a difficult problem to deal with in
the sciences. In the case of the invention of the telephone it took the form of
Bell patenting an invention apparently conceived during witnessing a
demonstration by Reis. In other strenuous cases, wily researchers have written
up papers based on material presented in lectures by others without any form of
acknowledgements. It can be hard to prove such things unless it can be traced
back to direct plagiarism.
The
Lysenko Affair
The Ukrainian Trofim Lysenko was an
agricultural experimentalist without proper scientific training. During the serious agricultural crisis in the 1930s, he impressed the
dictator in the former Soviet Union, Joseph Stalin, by wild claims that he had
discovered methods for growing crops without using fertilizers or minerals and
under unfavourable conditions (vernalization). Lysenko twisted ideas of Lamarck and Darwin and developed biological
theories in complete disagreement with theories by scientists outside the
Soviet Union, but his theories fitted well into Stalin’s political conceptions
that environmental factors rather than inheritance are decisive for growth and
achievements. Henceforward Stalin made Lysenko chief of all biological sciences
in the Soviet Union and Lysenko was awarded the highest signs of recognitions
for his ‘achievements’. The experiments conducted by Lysenko, among others
growing corn in Siberia, failed and caused very severe famine in Russia. Only after the death of Stalin in 1953 it was slowly officially
recognized that Lysenko was a fraud - a fact which had been known to good
Soviet biologists for years but could not be discussed openly during the Stalin
regime. A detailed account of Lysenko’s catastrophic impact on the biological
sciences in the former Soviet Union is given in [4].
Lysenko’s immunity to criticism within the
Soviet Union was first officially ended in 1964 after the dismissal of
Khrushchev - Stalin’s successor as Soviet leader - when the physicist Andrei
Sakharov spoke out against Lysenko in the General Assembly of the Academy of
Sciences and held him responsible for the poor state of Soviet biology and for
the defamation, firing, arrest, even death, of many genuine scientists. An
expert commission was appointed by the Academy to investigate the work of
Lysenko and a few months later, a devastating critique became public and the
Lysenko era finally faded out.
The Lysenko affair is a grave example of how a
political system can foster and support ‘research’ based on ideologies rather
than scientific methods. Lysenko collaborated eagerly with the system
persecuting the few scientists who opposed him. He was undoubtedly guilty of
scientific misconduct, even taking into consideration his poor scientific
training.
The
Concept of Scientific Misconduct
The first nation to take governmental steps
towards handling problems related to lack of integrity and honesty in the
sciences seems to be the USA, where offices for handling such problems were
installed in 1989 by two federal institutions, respectively the National
Science Foundation and the Public Health Service. The large majority of cases
have appeared within the area of responsibility of the Public Health Service,
where it is administratively handled by the Office of Research Integrity. This
office deals with fraud and misconduct in governmental funded biomedical and
behavioural research.
In 1996 the White House Office of Science and
Technology Policy took the initiative to phrase a definition of scientific
misconduct, which was published in the Federal Register on the 14th of October 1999 and later approved and adopted by the US National Academies of
Sciences [5].
The
Definition approved by the US National Academies of Sciences
The definition of scientific misconduct
(‘research misconduct’) currently being used in the USA is for good reasons
known as the FFP-definition (Fabrication, Falsification, and Plagiarism).
It goes as follows.
“Research misconduct is defined as
fabrication, falsification, or plagiarism in proposing, performing, or
reviewing research, or in reporting research results.
Fabrication is making up results and recording
or reporting them.
Falsification is manipulating research,
materials, equipment, or processes, or changing or omitting data or results
such that the research is not accurately represented in the research record.
Plagiarism is the appropriation of another
person’s ideas, processes, results, or words without giving appropriate credit,
including those obtained through confidential review of other’s research
proposals and manuscripts.
Research misconduct does not include honest
error or honest differences of opinions.”
The definition of scientific misconduct varies
slightly from nation to nation. The relatively narrow definition used in the
USA is a common core in all definitions of scientific misconduct.
I shall now give a short introduction to the
Danish system for handling misconduct in the sciences, including the definition
of misconduct used by the Danish Committees on Scientific Dishonesty (DCSD).
The
Danish Committees on Scientific Dishonesty
There are three committees in the Danish
system to deal with misconduct in research, one for each of the broad areas
“Health and Medical Science”, “Natural, Technological and Production Science”
and “Cultural and Social Science”.
The three committees, called the Danish
Committees on Scientific Dishonesty, have a joint secretariat in the Danish
Agency for Science, Technology and Innovation and they are chaired by a joint
chairman - a High Court Judge.
The committees were created in 1998. They
acted under “Executive Order no. 933 of 15 December 1998” until 1 August, 2005,
where a new set of regulations “Executive Order no. 668 of June 28, 2005” came
into force. The information about the committees in this article relates to the
new regulations [6].
Each of the three committees has 6 recognized
researchers as members and 6 alternates.
Each complaint made to the DCSD about
misconduct in research shall be considered by the committee under which the
research field of the defendant belongs – the decision about the relevant
committee is made by the chairman.
When a complaint is brought forward to DCSD,
all of the 36 members and alternates are informed about the complaint, and when
the relevant committee has made its verdict all 36 members and alternates are
again informed about the outcome.
The Danish system for handling misconduct in
the sciences is a comprehensive and serious one and it is not a playground for
escalating personal animosity among colleagues. Personal conflicts founded in
other things than research can only be handled at the place of work.
The definition of scientific dishonesty –
another phrase for scientific misconduct – used in Denmark at present is
slightly broader than the FFP-definition.
The
Definition used by the DCSD
“Scientific dishonesty shall mean intentional
or grossly negligent conduct in the form of falsification, plagiarism,
non-disclosure or any similar conduct involving undue misrepresentation of a
person’s own scientific work and/or scientific results. Included hereunder are:
- undisclosed fabrication and construction of
data or substitution with fictitious data;
- undisclosed selective or surreptitious
discarding of a person’s own undesired results;
- undisclosed unusual and misleading use of
statistical methods;
- undisclosed biased or distorted interpretation
of a person’s own results and conclusions;
- plagiarism of other person’s results or
publications;
- a false credit given to the author or authors,
misrepresentation of title or workplace;
- submission of incorrect information about
scientific qualifications.”
Adjusting
the handling of Misconduct in the Sciences
Scientists have always been dependent on
finding sponsors for their research. Nowadays the competition for funding is
severe and temptations for dishonesty in the way research is conducted thereby
become great.
A decisive incidence for bringing misconduct
in the sciences to the attention of the public and leading to the creation of
offices to handle asserted misconduct in a proper way was a famous case in the
USA.
The
Baltimore Case
In 1991 the prominent biomedical researcher
David Baltimore, Nobel Laureate 1975, was forced to resign from his position as
president of Rockefeller University, a position he had held since 1989. The
background was an incident going back to 1986 while he was director of a famous
biomedical research centre at Massachusetts Institute of Technology. In April
1986 David Baltimore with five co-authors published a paper on certain
reactions induced by transmission of genetic material from one strain of mice
to another strain. In May 1986 one of the co-authors, Theresa Imanishi-Kari,
was accused of fraud in reporting data on the genetic material. During
subsequent congressional hearings on the case, Theresa Imanishi-Kari was
vigorously defended by David Baltimore, who insisted on the right of scientists
to be the only judges of their work. The Congress was, however, of a different
opinion and took during the course of the trial initiative to establish the
federal agency Office of Research Integrity mentioned earlier. The Baltimore
case was concluded in 1996 with the complete acquittal of Imanishi-Kari,
whereas Baltimore, who received severe criticism in the first half of the
trial, had to resign his presidency of Rockefeller University in 1991. He
received rehabilitation in 1997 when he was nominated as president of
California Institute of Technology.
The Baltimore case has been instrumental for
raising the awareness of scientific misconduct in the sciences, particularly
the health sciences, and it fully demonstrated to the public how complicated it
can be to unravel fraud in the sciences without appropriate means. A detailed
and comprehensive description of the Baltimore case has been given by the
historian of science Daniel J. Kevles in the book [7], which was excellently
reviewed in [8].
A momentarily backlash to work for maintaining
good conduct in the sciences happened a few years ago in Denmark. The case
attracted considerably international attention.
The
Lomborg Case
The case in question was an accusation against
the Danish social scientist Bjorn Lomborg for scientific dishonesty in his book
“The Sceptical Environmentalist” [9]. The book argues and seeks to prove that
the main environmental problems created by mankind, like air pollution, are not
so serious after all and that it is better to use money on other problems for
mankind, like poverty and clean water. The accusation was brought for the
Danish Committees on Scientific Dishonesty early in 2002, and since the
complaints generally included areas of natural, social and health science, it
was decided to treat the complaints at joint meetings common to all of its
three committees. In its ruling in January 2003, the committees found that, by
customary scientific standards, Lomborg had acted at odds with good scientific
practice in his systematically one-sided choice of data and in his arguments.
If the book should be considered as science and not as a contribution to the
general debate, then in addition the scientific message had been so distorted
that the objective criteria for establishing scientific dishonesty had been
met. The committees did not find a sufficient basis, however, on which to
establish that Lomborg had misled his readers with intent or gross negligence.
Lomborg appealed the decision to the Danish
Ministry of Science, Technology and Innovation. The Ministry’s ruling in
December 2003 was critical of the committees work in a number of legal aspects,
including the rationale for regarding the book as a scientific work (research),
and also the use by the committees of the concept of ‘good scientific practice’
was severely criticised. Nevertheless, the Ministry let it be up to the
committees to decide whether to reopen the case. For legal reasons, this was,
however, not possible, since the committees found it unlikely that a new
examination would lead to substantial changes in the original ruling, which de
facto acquitted Lomborg of scientific dishonesty. Thus the Ministry’s ruling in
reality reflected an overruling of the decision by the Danish Committees on
Scientific Dishonesty. In several ways, the Lomborg case therefore ended
‘dangling in mid- air’.
The Lomborg case reveals the great problems
encountered in reaching balanced decisions when judging scientific conduct in cross-disciplinary
research. This is particularly true when political sciences are involved and
the line between research and debate is not clear cut – the problem with facts
and their interpretations.
As a political consequence of the Lomborg
case, the Danish legislature passed a new law “Executive Order no. 668 of June
28, 2005” for the Danish Committees on Scientific Dishonesty, which came into
force 1 August, 2005.
Some
Remarks on the new Regulations for DCSD
The wording of the definition of scientific dishonesty
as such is not dramatically changed in the new set of regulations for the
committees. There are, however, a few changes in the regulations for the
working of the committees, which could turn out to cause problems to various
degrees.
In the previous set of regulations, anybody
who came across a case of scientific dishonesty could bring it for the DCSD. In
the new regulations, the committees can only accept to investigate an
accusation of scientific dishonesty, if the whistleblower is a party, i.e. the
whistleblower shall have a direct personal interest in the case. It is,
however, possible for DCSD to take up a case of scientific dishonesty on its
own initiative if the case is of general interest for Danish science. It
remains to be seen whether this change will be an obstacle for taking up
serious cases of scientific dishonesty in Denmark in the future.
A change, which could turn out to be
troublesome, is that in the new regulations only people having an academic
training (candidate level at least) can be convicted for scientific misconduct.
Presumably, this will imply that e.g. medical nurses cannot be held responsible
for scientific misconduct, only medical doctors or scientists involved in
medical experiments.
A serious problem is that only researchers in
public institutions (universities, hospitals etc.) can be brought for DCSD in
cases of asserted scientific misconduct. A researcher in a private company is
exempted from being investigated by the DCSD unless the company chooses to
allow an investigation.
Some
Remarks on how the DCSD work
A case in Denmark, presently under
investigation, concerns a complaint against a Danish dentist. Following the
usual procedure, DCSD has advised people involved in the case not to go public
with details as long as the case is being investigated. Nevertheless, the case
was made public in an article in Jyllands-Posten September 17, 2006.
Unfortunately, some misunderstandings about how the DCSD work were brought
forward in the article.
Following the rules of DCSD, the case will be
handled by the “Committee for Health and Medical Science”, i.e. by 6 members or
alternates from that committee. To reach a verdict, at least 4 members and the
chairman (high court judge) shall be present. In Jyllands-Posten it was wrongly
asserted that the case would be handled by a small subcommittee of 3 arbitrary
researchers from the DCSD.
All verdicts in cases brought for the DCSD are
decided by researchers in subject areas close to the subject area of the particular
case. For help in investigating a case, the DCSD may appoint ad hoc committees
of experts for clarifying and elucidating the allegation about misconduct but
the final decision is made by members of DCSD as described.
Ethics
and Publication
Worldwide researchers are faced with constant
and increasing pressures by governments and universities for publication in
high prestige research journals. For many researchers it has become a goal in
itself to have a high citation count rather than thinking about making
contributions to knowledge.
The merits in presenting new research in a
scholarly way in carefully documented and comprehensive works, like monographs
and long memoirs, are not appreciated by those administrators who can count the
number of your publications but do not have time for assessing their quality.
Finding the ‘least publishable unit’ and slicing your work accordingly
therefore becomes a goal for many researchers.
The habit of slicing up publications in small
units has become a major concern to some scientific societies. ‘Salami
publication’ is not scientific misconduct as such but it presents a threat to
scholarship and knowledge by making it very difficult to trace the origins of
ideas and to get connected pictures of new theories.
Double or even multiple publication of almost
the same article in various scientific media is another bad habit resulting
from pressures for building up long publication lists. Double publication
could, and maybe even should, be considered scientific misconduct since it
blurs the scientific qualifications of researchers.
A particularly serious and growing problem in
the sciences is ‘multi-authored publications’ where some of the authors on a
publication did not contribute to its generation in any essential way,
sometimes not at all. In cases where a project team is involved in conducting
large scale experiments you often find hundreds of persons listed as authors of
a publication, including technicians who only made standard measurements.
Authorship should be considered a serious thing.
The Role
of Science Journalists in unveiling Scientific Fraud
Serious kinds of scientific misconduct include
as we have discussed fraud, fabrication of data, or extensive plagiarism, the
FFP-definition.
Accusations of scientific misconduct are
unpleasant for all the people involved. Obviously it is painful for the accused
but it can also turn into a nightmare for an informer whose motives are
violently contested by people seeking revenge. Spectacular cases of scientific
misconduct catch the interest of the media and they have serious consequences
for the careers of the people involved.
The Baltimore case showed that journalists
hunting for crooks, victims and heroes can do a lot of harm in complicated
cases involving questionable conduct of scientists. The case lasted for ten
years, and in the first five years so many things had been mixed up by the
press that eventually people asked whether the persons involved did not have
any legal rights. This was the reason for installing offices in the USA to
handle cases of fraud in the sciences. By the end of the Baltimore case, the
roles of crooks and heroes had been reversed but all the scientists involved
were hurt and somehow became victims of the case. The consequences of wrong
premature conclusions are so devastating for the scientists involved in cases
of scientific fraud that journalists should be more than careful.
Science journalists certainly have a big role
when it comes to explaining in understandable terms and in a fair and neutral
way what a serious allegation of scientific misconduct is about, and reporting
about the outcome when the case is concluded. Journalists always have the
advantage over a legal system that they can reveal feelings and common sense
attitudes to cases, which the DCSD cannot.
In some of the examples of asserted scientific
misconduct we have discussed, strong political interests seem to have been at
stake. In such cases a free and independent press is, of course, absolutely
essential. It seems that some journalists understood the Lomborg case better
than some of the managing editors of leading Danish newspapers.
The role of the whistleblower in the sciences
should also be considered. It is important to support the colleagues who insist
on honesty in publications and experimental work at a time with a constant
pressure for high productivity. Here journalists can be of big help, not least
since a journalist can ask more questions than a researcher in the DCSD.
Journalists can also be of help in displaying bad publication habits.
A large majority of scientists strive for
maintaining good conduct in the sciences and are upset by serious cases of
scientific misconduct. It is important for the honest scientists to see that
improper behaviour is dealt with and not just neglected as is often the case.
As the competition for funding increases the number of cases of serious
misconduct in the sciences is likely to grow unless more emphasis is put on
education of the coming generations of scientists in ethical standards.
Constructive debates about such issues led by science journalist can
undoubtedly also be valuable.
References
- Broad,
W. and Wade, N., ‘Betrayers of the Truth’ (Simon and Schuster, New York, 1982).
- Martinson,
B.C., Anderson, M.S. and de Vries, R., ‘Scientists behaving badly’ (Nature 435 (2005), 737-8).
- Coe,
L., ‘The Telephone and its Several Inventors’ (McFarlane, 1995).
- Soyfer,
V.N., ‘Lysenko and the Tragedy of Soviet Science’ (Rutgers University Press,
1994).
- Goodman,
B., ‘New definition for misconduct a step closer’ (The Scientist 2000, Issue
14, 1-13).
- Executive
Order No. 668 of 28 June 2005 on the Danish Committees on Scientific
Dishonesty. Available at the website for the Danish Agency for Science,
Technology and Innovation: www.fist.dk
- Kevles,
D.J., ‘The Baltimore Case, A Trial of Politics, Science and Character’ (W.W.
Norton & Company, New York, London, 1998).
- Rehfeld,
J.F., ‘Baltimore-sagen: en lære om forskningens indre og ydre fjender’ (BioZoom
nr.2, 2000).
- Lomborg,
B., ‘The Skeptical Environmentalist: Measuring the Real State of the World’
(Cambridge University Press, 2001).
This article is based partly on the manuscript
for a lecture to the Danish Science Journalists’ Association on October 12,
2006, and partly on my article “Good conduct in the sciences”, published in the
Annual Report 2005 from the Danish Committees on Scientific Dishonesty. |
