《自然》：中国制定处理科研不端的新规针对大量制造假研究的论文工场
中国制定处理科研不端的新规针对大量制造假研究的论文工场
更严厉打击学术造假的措施看上去不错，但批评者称如何实施仍然是个问题
记者：Smriti Mallapaty
《自然》2020年8月21日
中国科技部发布至今最为全面的处理科研不端的规定。这些将在下个月生效的措施归纳了构成违规和做适当处罚的情形。它们适用于从事科技活动的任何人，包括研究人员、评审人员和机构领导。
该政策也首次包括了独立合同工的违规，例如那些出卖论文、伪造数据、替研究人员写稿或投稿。该规定是为了对付研究人员普遍使用被称为论文工场的公司的现象，这些公司常常通过伪造数据撰写论文。
有些科学家说，这些规定会有助于抑制坏行为并提高中国机构的科研道德水平。上海复旦大学研究科技政策的唐力说，它们“前进了一大步”。
但是其他人怀疑这些改变能有什么结果，因为关于不端行为的规定已经有了，但是没有执行。
“他们不需要制定新规。已经有足够多的旧规定可以用，”住在加州圣地亚哥的作家方舟子说，他致力于揭露中国的科学造假。
想要真正产生影响，中国政府需要拿某个高端案件作为例子，位于伊利诺州Peoria的布拉德利大学图书馆员陈晓天说。“应该让那些从事科研不端的作者承担后果，特别是那些位高权重者。”
中国在科研不端方面存在的问题为时已久，引起全球关注。案例涉及到贿赂、剽窃、伪造数据和捏造同行审稿，导致大量的论文被撤稿。
早自2006年以来中国已发布了众多处理这一问题的政策，但科学家们说由于不执行，局势恶化了。“目前的情况比十年前更严重，”方说。“不端行为已成为系统性的，而且商业化了。现在有很多论文写作公司在帮助研究人员撰写和发布假论文。”
2017年，在一次重大丑闻之后，中国科技部保证要打击科研不端。这个丑闻涉及发表在《肿瘤生物学》上的107篇论文被撤稿，该刊以前由斯普林格·自然出版。这些论文被撤稿是因为它们的审稿是伪造的，而且很多篇是由论文工场制造的。
方说，最新的规定就是科技部当时承诺的一个结果。根据这些规定，科研不端包括伪造结果、剽窃、未经伦理批准就做实验、干预同行评议和贪污科研基金。那些导致严重损害或财政损失的研究人员将得到更严厉的处罚。
处罚包括从警告到取消奖金、奖励和荣誉头衔，甚至暂时或永远禁止研究人员申请政府基金。报告自己违规或承认错误并试图纠正错误的研究人员将从宽处理。但是累犯、掩盖违规或恐吓举报者将受到严厉处置。
大连理工大学研究科研道德问题的研究人员王飞说，该政策也给予受调查的机构或个人上诉的权利，包括通过法院上诉，这对正当程序是重要的。
根据该政策，严重的违规必须公布。“要求透明将有助于防止不端行为，”李说。
李说，这些规定将会促使研究人员配合调查，以及机构对可能的不端行为做更深入的挖掘。但是措施如果要成功地打击坏行为的话，政府需要追究违规者的责任，李说，“接下去会怎么做才是关键。”
但是中国以往的记录表明这些最新规定对减少不端行为不会有什么效果，因为已有的规定并没能制止这类行为，方说。
“伪造数据、修改图片和伪造同行评议的做法被认为是可以的，把研究工作和撰写论文外包给论文工场被认为是可以，这已经成了文化，”陈说。例如，在2月，一些研究人员发现450多篇由中国医院的作者写的论文的图片有问题，这些研究人员说这些论文可能都是来自一家论文工场。
陈说，有些论文工场还在继续公开做广告拉生意，尽管在2015年的时候有几个政府部门发布规定禁止研究人员使用这些公司写论文和投稿、修改除了语法错误之外的论文内容或假造同行评议。陈说，政府需要继续打击这类服务。
中国科技部没有答复《自然》关于已有的处理不端行为条例是否正在实施的问题。
但是李注意到，这些条例已提高了中国的科研道德，虽然很难提供证据表明二者直接相关。有越来越多关于中国科研不端案例的报道，这可能只是反映了监督更严了，这个问题在中国和在国外更受重视了，她说。
（方舟子翻译）
nature
NEWS? 21 AUGUST 2020
China’s research-misconduct rules target ‘paper mills’ that churn out fake studies
Measures to crack down harder on falsified work look good on paper, but critics say that enforcement will continue to be a problem.
Smriti Mallapaty
China’s science ministry is set to introduce its most comprehensive rules so far for dealing with research misconduct. The measures, which come into effect next month, outline what constitute violations and appropriate punishments. They will apply to anyone engaged in science-and-technology activities, including researchers, reviewers and heads of institutions.
The policy also includes, for the first time, violations by independent contractors, such as those who sell academic papers, fabricate data and write or submit articles on behalf of researchers. The rule is designed to tackle researchers’ widespread use of companies known as paper mills, which produce manuscripts that are often based on falsified data.
Some scientists say the regulations will help to curb bad behaviour and improve research integrity in Chinese institutions. They are a “big step forward”, says Li Tang, who studies science policy at Fudan University in Shanghai.
But others doubt the changes will make a difference, because misconduct regulations already exist, but are not enforced.
“They don’t need to make new rules. There are plenty of old regulations ready,” says Shi-Min Fang, a writer based in San Diego, California, whose work focuses on exposing scientific fraud in China.
To have a real impact, the Chinese government needs to make an example of a high-profile case, says Xiaotian Chen, a librarian at Bradley University in Peoria, Illinois. “There should be consequences for authors with research misconducts, especially those in powerful and important positions.”
China has had a long-standing issue with research misconduct, which has drawn global attention. Cases have involved bribery, plagiarism, falsified data and forged peer review, and have led to a large number of article retractions.
Numerous policies were introduced as long ago as 2006 to address the problems, but scientists say that non-enforcement has only aggravated the situation. “The current situation is worse than a decade ago,” says Fang. “Misconduct has become systematic and commercialized. Now there are many paper-writing companies helping researchers write and publish fake papers.”
New rules
In 2017, China’s Ministry of Science and Technology (MOST) committed to cracking down on research misconduct in the wake of a major scandal. This involved the retraction of 107 research papers in the cancer journal Tumor Biology, previously published by Springer Nature. The articles were retracted because their reviews had been fabricated, and many papers had been produced by paper mills. (Nature is published by Springer Nature, and Nature’s news team is editorially independent of the publisher.)
The latest rules are a result of the ministry’s promise, says Fang. According to the measures, research misconduct includes falsifying results, plagiarism, running experiments without ethics approval, meddling in the peer-review process and embezzling research funds. Researchers whose actions cause severe harm or financial losses will be given harsher punishments.
Penalties can range from warnings to revoking bonuses, awards and honorary titles, and even banning researchers from applying for government funding, temporarily or permanently. Researchers who report their own violations or admit to mistakes and attempt to correct them will be given lighter punishments. But repeat offenders and those who cover up misconduct or intimidate whistle-blowers will be dealt with severely.
The policy also gives institutions or people under investigation the right of appeal, including through the courts, which is important for due process, says Wang Fei, a science-policy researcher at Dalian University of Technology who has written about research integrity.
Serious violations must also be made public, according to the policy. “This requirement for transparency will help to deter misconduct,” says Li.
Mixed reactions
Li says the rules will encourage researchers to cooperate in investigations, and institutions to dig deeper into potential misconduct, she says. But for the measures to successfully curb bad behaviour, the government will need to hold those who violate the rules to account, says Li. “What happens next is crucial.”
But China’s track record suggests that the latest rules will have no effect on reducing misconduct, because existing rules have failed to stop such behaviour, says Fang.
It has become “the culture that fake data, photoshopped images and fake peer reviews are OK, and that outsourcing research and writing to paper mills is OK”, says Chen. In February, for example, a group of researchers identified more than 450 papers with problematic images by authors affiliated with Chinese hospitals, which the researchers say probably came from a paper mill.
Chen says that some paper mills continue to advertise their services openly, despite several government agencies publishing rules in 2015 that prohibit researchers from using these firms to write and submit their manuscripts, revise content other than grammar, or provide false peer review. The government needs to crack down on these services, says Chen.
The MOST did not respond to Nature’s questions about whether existing misconduct regulations are being enforced.
But Li notes that regulations have improved research integrity in China, although it is difficult to point to evidence directly linking the two. Increased reporting of research misconduct cases from China could just reflect stricter policing, and that the issue is being taken more seriously in China and abroad, she says.