Nigel Praities hears some dissenting voices

Brilliant, blue-eyed physicist, Jan Henrick Schön, was poised to become the youngest ever Co-director of the prestigious Max Planck Institute before his dubious methods were brought to light. Although regarded as one of the most promising and prolific researchers of his generation – publishing about 80 papers on nanoelectronics in two years – he was found to have falsified and fabricated data by an independent committee, and was fired by Bell Laboratories in 2002.

While this audacious act of fraud shocked the physical science community, there was little evidence of any soul-searching to prevent this happening again. Rather, Schön was presented as a tragic, misguided figure; but not representative of the state of science itself. Yet these events, and the even higher profile fall of cell biologist Woo Suk Hwang in Korea last year, have raised questions about how widespread misconduct is in science and whether we can trust scientists to behave ethically.
The UK government has sought to address this question, with a proposal from the Council for Science and Technology for a new Universal Code for Scientists.1 This code has been largely welcomed by the scientific community, but some have voiced concern that it is too simplistic and is unlikely to be effective.

A ‘useful framework’

The government’s code broadly covers three main areas: research ethics, reflection among scientists on the implications of their work and communication with the public. The code is designed to be a basis for discussion and to ‘raise awareness’ of ethical issues in science. 

In the increasingly competitive environment of science, can this code address the considerable pressures on scientists to get results?

‘The public increasingly demands that scientific developments are ethical and serve the wider public good, as evidenced by the debate on stem cell research,’ said Sir David King. ‘The code has been developed in my office to help us meet this challenge.’ In March of this year, as part of National Science and Engineering Week, King issued a challenge for scientists to adopt and help promote the code.
The code is regarded as a ‘useful framework’ for scientists to develop their own codes and ‘has considerable value as an educational and awareness-raising tool to ensure that scientists are reminded of their legal and ethical responsibilities,’ said Professor Martin Taylor, Vice President of the Royal Society. In addition, the Royal Society of Engineers has already used the code to develop a Statement of Ethical Principles for professional engineers.2
As a research scientist in the Department of Physiology at University College London, Dr Frances Edwards regards the code as ‘common-sense: intrinsic things that make good science. But I can’t imagine myself being in the situation where I would say “Shall I do this or not? I had better look at the code!”’ She thinks that it is more useful for undergraduate science students and the public to understand that these principles are ‘part of your training and your general personal ethics as a scientist’.
King also believes that the code can be used to guide young researchers and, in the longer term, he would like to see it included in the school curriculum, thereby helping to instil in young people ‘the values embedded in the code long before they enter the workplace’. However, is it naïve to assume that these principles are second-nature to scientists? In the increasingly competitive environment of science, can this code address the considerable pressures on scientists to get results and publish, conditions blamed for both Hwang and Schön’s unethical behaviour?

Misconduct in science

‘The code suggests that misconduct occurs simply because people are naughty. But basically people are in an equilibrium with their environment,’ says Dr Ronán Conroy who is an epidemiologist and member of the Ethical Committee at the Royal College of Surgeons in Ireland. ‘In an environment which is being increasingly driven by commercial pressures and vast sums of money to be won or lost, it is no longer a cricket pitch.’
The amount of misconduct in science is hard to measure, but one recent study of 3,247 early and mid-career researchers showed that that one in three admitted committing some form of misconduct during the previous three years.3
‘I think that the corruption in science is unwitting rather than witting, but I do think that it is widespread,’ says Conroy. ‘For example, the sidetracking of medicine by the pharmaceutical industry, and the massive sums of money spent on cosying up to the medical profession, compromise the integrity of medical research.’ 

A recent review of 342 clinical trials showed that reported outcomes were significantly more likely to be positive if funded by for-profit organisations

Clinical trials funded by pharmaceutical companies are better funded and designed, but are often strategically positioned to show a particular drug in a better light. A recent review of 342 clinical trials showed that reported outcomes were significantly more likely to be positive if funded by for-profit organisations.4 These subtle, corrupting influences are much harder to identify than the blatant fraud of Schön or Hwang, and are not addressed by the new code.

Enforcement of the code

Known as the ‘pride of Korea’, Hwang was a respected scientist and a national hero in his homeland in 2004. But the whiff of unethical behaviour around Hwang’s practices led to the discovery of the greater stench of fraud in his closet. Allegations that his female researchers had been coerced to donate eggs for research were published in the journal Nature, leading to further revelations that staff were encouraged to counterfeit results and that his data on cloned human embryos were fake.
A written code of conduct for scientists may provide greater encouragement for researchers in a similar position who wish to whistle-blow and report dubious lab practices. Yet, King’s code is entirely voluntary and there is no mention of how to encourage good behaviour in science.
Professor Steve Fuller, a sociologist at the University of Warwick, argues that this lack of teeth means that the code is unlikely to have an effect on scientific misconduct. ‘King’s code is just a paper tiger,’ says Fuller. ‘If you had a serious code you would define misconduct in operational terms. You could then identify who had committed the offence and do something about it.’
Including stringent definitions of misconduct, such as the submission of fabricated or duplicated data, would enable greater clarity and could enable the code to be policed by a public body. This is common in other areas of society which are self-regulated, for example the Press Complaints Commission for the media, or the Association of the British Pharmaceutical Industry for drugs companies.
However, it could be debated how effective this type of control is in curbing bad behaviour or building public trust in the media or the pharmaceutical industry. Fuller admits that this approach has a major drawback in the case of science: ‘If you have a real, professional code with sanctions, it could end up removing trust from science.’

A PR gimmick?

Perhaps the problem with the code is that it is more a list of value statements than a code of conduct. It assumes high-profile cases of fraud, such as those committed by Schön or Hwang, are the result of science going wrong, rather than a product of the way science works itself. ‘It looks to me like a PR gimmick,’ says Fuller. ‘I think that King probably thinks that we are fine and the problems with science are outside of it.’

The code assumes that high-profile cases of fraud are the result of science going wrong, rather than a product of the way science works

It is reasonable to assume that most scientists are honest, but Schön and Hwang highlighted some cracks in the august façade of science. They exposed the pressure to publish results and the weaknesses of the peer review process as a regulatory mechanism. These may reflect more fundamental flaws in the structure of modern science. Papering over these cracks with some weak value statements is unlikely to stop them from appearing again in the future.
The privileged position of science in our society depends on continued public trust in scientists and their work. This remains at a high level, with 70 per cent of the public trusting that scientists tell the truth, compared with 23 per cent in the case of government ministers. But it is debatable whether this new code for science is the tool to bolster this trust and prevent misconduct. That largely depends on whether Schön or Hwang really were two bad apples, or signs of a wider disease within the heart of science.

1. 1. Department of Trade and Industry. Universal Ethical Code for Scientists. See http://tinyurl.com/3dvyeu
2. 2. Royal Society of Engineers, Statement of Ethical Principles. See www.raeng.org.uk/news/releases/pdf/ rae_statement_of_ethical_principles.pdf
3. 3. Martinson et al. (2005), Scientists behaving badly. Nature; 9, 435 (7043): 737–8
4. 4. Ridker PM and Torres J. (2006), Reported outcomes in major cardiovascular trials funded by for-profit and not-for-profit organisations: 2000-2005. JAMA; 295: 2270-2274

Nigel Praities is currently completing an MSc in Science Communication at Imperial College London